3. CLINICAL PRACTICE MODULE 3.1 Clinical Practice Case Studies 3.1.1 Missed Diagnosis by Another Healthcare Provider BRIEF DESCRIPTION Mdm CAH is a 76 year old Chinese lady with a medical history of diabetes mellitus, hypertension and hyperlipidaemia. She was admitted to the orthopaedic department of a restructured hospital after slipping and falling in her bathroom due to a slippery floor. She sustained a right distal femur fracture which was fixed. Mdm CAH was subsequently discharged to Ang Mo Kio - Thye Hua Kwan Hospital (AMKH) for rehabilitation with instruction to allow full weight bearing on the right lower limb. On the day after admission, Mdm CAH complained of right ankle pain and swelling which she said was present since the operation. According to her, she told the doctor at the restructured hospital about it but it was not investigated. She denied any fall or trauma post operation. Examination of the right ankle revealed mild swelling and tenderness over the lateral malleolus with pain on passive range of movement. The neurovascular status was intact distally. An x-ray of the joint showed an undisplaced lateral malleolar fracture. As AMKH has no facilities for application of a backslab or cast, the patient was transferred to the Accident and Emergency department of the restructured hospital which she was discharged from for further management. The patient was admitted and treated conservatively with a cast. She was subsequently transferred back to AMKH for continued rehabilitation, but this time she was not allowed to weight bear on the right lower limb for 6 weeks. During her second admission to the restructured hospital, the family sought to have her hospitalisation bills and transportation fees waived as, in their opinion, the right ankle fracture should have been picked up and managed during the first admission and thus avoiding this second admission. However, according to the patient’s son, the restructured hospital stated that examination of the right ankle was normal during the first admission and they denied any record of the patient complaining of pain in that joint. The case has been referred to their Clinical Services department. Mdm CAH’s family approached us with two questions: 1) Could they have a copy of the discharge summary from the first admission to the restructured hospital? 2) Should the doctors have picked up the fracture during the first admission? 7

Providing a copy of the discharge summary The family wanted these documents to see if anything was documented regarding the right ankle. Their contention is that the patient has the right to have a copy of these documents as it is her information. Generally, the doctor’s copy of the discharge summary contains a synopsis of the patient’s admission including diagnosis, investigations and management. It serves as a means to facilitate continuity of care as the patient transits from one healthcare provider to another. It could be seen as a private correspondence between healthcare providers for the purpose of the patient’s management. A wholesale duplication of the discharge summary could be deemed a breach of copyright as there is no implicit permission given to duplicate the document other than for the patient’s management. Moreover, the discharge summary clearly indicated in the header that it was not to be given to the patient. Hence, it is quite evident that the discharge summary cannot be given to the patient’s family, even if the patient consented to the release of her medical information to them. However, the patient has a right to her medical information, although the document itself belongs to the restructured hospital. Hence, we could inform the family, with the patient’s permission, whether anything pertaining to the right ankle was documented and if so, what was documented about it. In any case, there was nothing about that joint which was documented. The family was also referred to the restructured hospital’s Clinical Services department should they still want a duplicate of the discharge summary. Alternatively, or if the restructured hospital is reluctant to furnish them with any documents, the family could get a lawyer to write in for discovery of documents as a prelude to trial. Opinion on whether the doctors at the restructured hospital should have picked up the fracture It would seem on the face of it that if Mdm CAH told the doctor of the right ankle pain, it should have been pursued and investigated, like how it was done when the patient was admitted to AMKH. One can also understand how a reported symptom maybe considered minor and therefore not followed up upon in the context of an acute hospital where there are other more pressing matters at hand. However, it is difficult to comment on culpability of the other doctors due to a few factors:

8

1) Incomplete facts We do not know whether Mdm CAH did indeed highlight the pain in the right ankle during her first admission. Moreover, we do not know whether the joint was examined and what the findings were and whether the resultant Ottawa Ankle rules score would have prompted an x-ray of the right ankle to be ordered. It could be that unlike what was noted at AMKH, there was no tenderness or swelling at that point in time. 2) Commenting on the management of a specialist not in the same field While Family Physicians have to some extent a working knowledge of orthopaedics, we may nonetheless lack sufficient competency in that specialist field to offer a reasonable comment. Moreover, whether the orthopaedic doctors were negligent should be measured against the standard of a group of orthopaedic doctors of similar seniority per the Bolam Principle. Offering an opinion to the family would give the family the false impression that I was an appropriate “expert witness” when I was not. In addition, the SMC Ethical Code and Ethical Guidelines stipulates that “a doctor shall refrain from making gratuitous and unsustainable comments which, whether expressly or by implication, set out to undermine the trust in a professional colleague’s knowledge or skills”. As the patient would continue to be followed up at that restructured hospital, it was important for her continued management that I do not inadvertently or otherwise further jeopardise the already soured doctor-patient relationship. Moreover, making baseless comments negatively impacting the reputation of another healthcare professional would also be considered defamation. I was also acutely aware of my ethical duty to be truthful in my statements and not to cover up any obvious errors (ethical principle of veracity). Moreover, in my choice of words and tone, I needed to avoid being too defensive or uncooperative with their queries, lest I give the family the impression that I was more concerned about protecting those from my own profession than helping the patient. Hence, I explained that it is not possible for me to offer an opinion due to my lack of knowledge of what transpired at the restructured hospital as there were many factors at play (e.g. symptoms reported, examination findings, etc). I also did not have the requisite specialist knowledge. I referred the family instead to discuss this with their Clinical Services department, but should they need any assistance in understanding what usually happens as part of the complaints process or any medical jargon or procedures they may not understand, they could approach me (principle of fidelity – being there for the patient and being her advocate). They were generally satisfied with my response.

9

LEARNING POINTS •

The handling of medical correspondence between healthcare professionals should be done sensitively and with the awareness of the medico-legal implications.

•

When an opinion on the management of care of another healthcare professional is being sought, it is important to refrain from making disparaging remarks yet adhere to the principle of veracity – being truthful and not cover up the errors of others. This would promote the quality of the doctor-patient relationship and avoid disrepute to the profession as a whole.

•

Whatever the situation, the Family Physician has to uphold the principle of fidelity – being there and being an advocate for the patient – to help them navigate through the complex world of medicine with its terminology and processes opaque to lay people.

•

It is also important to cultivate an environment of collegiality between specialists and general practitioners so that patients can transit smoothly between specialist and primary care and where trust is built up between patients and whichever doctor is attending to them. This all translates to better clinical outcomes for patients, a reduction in unwarranted complaints and a more satisfying doctor-patient relationship.

•

Such collegiality can be facilitated through greater dialogue between specialists and general practitioners (e.g. at CME events, regular meetings on cases which are comanaged, etc). Medical students should also be taught to see healthcare provision as a team effort where no one provider is seen as better than the other. To this end, we have to watch how we teach medical students and clinical scenarios beginning with how the general practitioner missed a diagnosis but is picked up by the specialist who saves the day (or vice versa) are best avoided (Kamien 2005).

REFERENCES Bolam v Friern Hospital Management Committee [1957] 1 WLR 582. SMC Ethical Code and Ethical Guidelines, paragraph 4.3.3. Chin Jing Jih. Collegiality. SMA News. Jan 2013; 10-11. Max Kamien. Communication and courtesy between medical professionals. MJA. 2005; 183(11/12):629.

10

3.1.2 Long Term Follow-up of Down Syndrome Patient BRIEF DESCRIPTION Mr TCT is a 28 year old Chinese gentleman with a history of Down Syndrome. He was premorbidly ADL independent and community ambulant. He slipped on the wet floor in his bathroom at home after a shower and landed on his back. There were no antecedent symptoms of giddiness or weakness, and he was able to stand up and ambulate post fall. However, due to the back pain, his mother brought him to Parkway East Hospital for further assessment where a L1 compression fracture was discovered. He subsequently underwent a kyphoplasty and was allowed to fully weight bear post-operation. He was transferred to AMKH for rehabilitation. While at AMKH, further history from his mother revealed that Mr TCT was previously seen by a paediatrician when he was much younger for routine follow-ups but he had no major complications (including cardiac complications) of Down Syndrome. He has not seen any doctor regularly for many years since, despite the fact that Down Syndrome adults are prone to various medical problems. Through screening, the following medical problems related to Down Syndrome were uncovered: 1) Depression Mr TCT appeared to have low mood. He had poor eye contact and a monotonous voice, but what was crucial was the further history from his mother which revealed that he has been having low mood with anhedonia, lethargy and hypersomnia for more than 1 year, possibly triggered when he left his job at a charitable organisation. His mother was encouraged to contact the Down Syndrome Association after Mr TCT regained more of his function to see if he could participate in their Adult Enhancement Program so as to improve his employability. Mr TCT was also given Escitalopram 10mg ON to aid with his mood and a follow-up appointment with the psychiatrist was made. 2) Osteoporosis Although Mr TCT is relatively young, a dual-energy x-ray absorptiometry (DEXA) scan was done to assess his bone mineral density in view of the history of a L1 compression fracture and Down Syndrome. It revealed a T score of -5.2 SD for the lumbar spine and -3.8 SD for the hip, indicative of osteoporosis. His creatinine and calcium levels were normal but his 25OH Vitamin D level was low at 11ug/L. Vitamin D3 replacement was started and his Vitamin

11

D level subsequently increased to sufficient levels (20 ug/L) after about a month. He was then started on Alendronate 70mg once a week. 3) Obesity Mr TCT’s BMI was 29 kg/m2 which placed him in the high risk category. According to his mother, Mr TCT is indiscriminate when it comes to food and enjoys eating high fat fast food like KFC. He also does not do any exercise. His lipid profile and glucose levels were, however, normal. Mr TCT did not have symptoms of obstructive sleep apnoea or osteoarthritis of the knees. Dietary advice was given to Mr TCT as well as his mother who decides on his meals. After his functional recovery, Mr TCT would need to be more physically active, and participation in the Adult Enhancement Program as mentioned above would be one initial step which should be taken. 4) Subclinical hypothyroidism A thyroid function test showed normal free T4 of 18 pmol/L and a raised TSH of 6.77 mIU/L (upper limit is 4.2). Mr TCT did not have any symptoms of hypothyroidism apart from the depression and lethargy. He also had no goitre. It was difficult to determine whether Mr TCT should be given thyroxine replacement at this stage as there is scant evidence on the risks and benefits of treating subclinical hypothyroidism in the Down Syndrome population. On the one hand, the subclinical hypothyroidism may be contributing to his depression and obesity. Yet his depression was more likely to be triggered by the loss of his job and his recent illness. His obesity also seems to be due more to lifestyle causes. Hence, treatment with thyroxine may not result in much benefit. Moreover, if thyroid was replaced excessively, this would hasten the rate of bone turnover and worsen his osteoporosis. Thyroxine replacement was therefore withheld for the time being and a memo was given to the general practitioner / polyclinic doctor to recheck the thyroid function test in 2 months. If normal, Mr TCT would need regular checks of his thyroid function at least once every 2 years (Henderson et al 2007). Screening for other medical problems Screening of his eyes did not reveal any cataracts and his hearing was normal. Heart sounds were normal and therefore not suggestive of any cardiac defects like an atrial or ventricular septal defect or mitral valve prolapse. His abbreviated mental test (AMT) was 9/10 which was not suggestive of dementia which is in line with the mean age of Alzheimer’s dementia in Down Syndrome adults being 52 years old (range from 40 to 61 years old (Henderson et al 2007). Although coeliac disease is associated with Down Syndrome and

12

Vitamin D deficiency may suggest malabsorption of fat-soluble vitamins which is also associated with coeliac disease, Mr TCT did not have gastrointestinal symptoms such as abdominal pain, steatorrhoea, constipation or diarrhoea to warrant serological screening. Screening for social issues Mr TCT has an older brother who is unmarried and currently employed and staying together with him and his parents in a three storey terrace house. His mother did not report any poor family dynamics with respect to his brother. The family also has a domestic helper who assists with the household chores. His mother, a homemaker, has been his main caregiver. She did not voice any caregiver stress as the patient has been relatively well prior to this illness, nor did she express any concerns about Mr TCT’s care should she one day be unable to care for him. Nonetheless, I advised the mother to contact the Down Syndrome Association who can refer her to the Special Needs Trust Company which offers subsidised trust services including writing of a will and identifying his brother, for example, as his legal guardian, as well as case management services. LEARNING POINTS •

Adult Down Syndrome patients are at higher risk of certain treatable medical conditions. However, some patients do not have a principal physician whose responsibility is to provide the necessary on-going medical surveillance and as such, these medical conditions may not be detected and treated on a timely basis, resulting in morbidity and a poorer quality of life.

•

There is currently no medical speciality dedicated to the care of adults with Down Syndrome. As part of providing preventive care, the family physician would actually be best placed to conduct such regular surveillance and provide the coordination required to match the patient with the appropriate level of specialist care. However, this would require the family physician to be aware of the health needs of this special group of patients and to draw up a system of regular review.

•

A system of adequate handover when the patient crosses from the paediatric to the adult age group would be essential to ensure continuity of care. However, until such a system is established, it falls on the family physician to be alert to the need for regular surveillance of adults with Down Syndrome and flag them out on an opportunistic basis as in this case.

•

Evidence based guidelines detailing the necessary health checks and screening (especially frequency) would also assist primary care physicians in caring for adults with special needs. It would also standardize care for such patients and this would open the

13

possibility for non-medical healthcare professionals to play a role in the health surveillance of such patients. An example of such a guideline would be those of the American College of Preventive Medicine. However, even with such guidelines in America, a recent study found inconsistent preventive care in Down Syndrome patients. REFERENCES Alex Henderson, Sally A Lynch, Steve Wilkinson, Morag Hunter. Adults with Down’s Syndrome: the prevalence of complications and health care in the community. Br J Gen Pract. 2007 Jan;57(534):50-5. Scanlon SA, Murray JA. Update on celiac disease—etiology, differential diagnosis, drug targets, and management advances. Clinical and Experimental Gastroenterology. 2011;4:297–311. Leigh Wilson. Preventive Care for Adults With Down Syndrome. Medscape 2010: http://www.medscape.org/viewarticle/715382. Jensen KM, Taylor LC, Davis MM. Primary care for adults with Down syndrome: adherence to preventive healthcare recommendations. J Intellect Disabil Res. 2013 May;57(5):409-21.

14

3.1.3 Difficult to Manage Fall / Fracture Risks BRIEF DESCRIPTION Mr TJK is an 83 year old Chinese gentleman with a medical history of: 1) Diabetes mellitus complicated by nephropathy (CKD 3) and cataracts, on Metformin 850mg BD and Tolbutamide 200mg BD. 2) Hypertension on Losartan 100mg OM and Bisoprolol 2.5mg OM. 3) Hyperlipidemia on Lovastatin 10mg ON. 4) Ischemic heart disease with previous percutaneous intervention in 1993 and 2007 (bare metal stent inserted) and coronary artery bypass graft in 1995. On Aspirin 100mg OM and Omeprazole 20mg BD. 5) Obstructive sleep apnea on CPAP. 6) Parkinson's disease on Madopar 125mg TDS. 7) Antral and incisural gastritis on OGD done 2011, recent haemoglobin normal. 8) Constipation with normal colonoscopy in 2010, on laxatives. 9) Mild cognitive impairment diagnosed since 2011, MRI brain then showing marked cerebral involution. 10) Allergy to penicillins and adverse drug reaction to hydrochlorothiazide (hyponatremia). Premorbidly, he was independent in the activities of daily living and was community ambulant. He was advised to use a walking stick but he has not been using one. He has bilateral hearing impairment and wears hearing aids. He lives with his wife in a 2 storey terrace house with the bedrooms located on the second level. His two children live overseas. A cleaner comes once a week to help with the housework. Mr TJK was admitted to Tan Tock Seng Hospital after a fall (first episode) while climbing up the stairs at night after he missed a grip on the handrail. He hit his head on the stairs but did not lose consciousness or had any headache, nausea or vomiting. He did however complain of a lower back pain but was able to ambulate by himself post fall. There was no antecedent giddiness, chest pain, shortness of breath or palpitations. He was noted to have a scalp hematoma with no intracranial bleed or infarct on CT and an x-ray of the lumbar spine revealed a compression fracture of the T9 vertebra with

15

osteoporotic bones and severe facet joint degeneration at L5-S1 and S1-S2 levels. Sodium levels were also noted to be low at 128 likely secondary to SIADH. While his blood pressure was noted to be high (175/75, up to a peak of SBP 200), no postural drop was noted. ECG did not show any rhythm abnormalities. He was given codeine phosphate for low back pain but developed delirium which resolved after it was stopped. He was fluid restricted and his anti-hypertensive medications were titrated upwards for better blood pressure control. Mr TJK was subsequently transferred to AMKH for slow stream rehabilitation. Medical Problems and Management 1) Recent fall with scalp hematoma The fall was precipitated by environmental factors (poor lighting, stairs), failure to use appropriate walking aids and hyponatremia. The predisposing factors of the fall included impaired vision (due to cataracts) and hearing, Parkinson’s disease, decreased proprioception (due to diabetes mellitus) and cognitive impairment (AMT 7/10). Hyponatremia was corrected after fluid restriction of 1.2L a day. Patient was scheduled for a cataract operation in December 2012. Madopar was increased to 187.5mg TDS to reduce rigidity and tremors. Advice was given regarding sufficient lighting, non-slip flooring and appropriate stair railings. Safety awareness was reinforced during therapy sessions, including consistent use of walking aids. However, Mr TJK did not like using walking aids as it made him feel like a disabled person, especially since he was a high functioning individual previously (Resnik et al 2009). It was a challenge therefore to help him recognise the importance of the use of walking aids, yet affirming his dignity at the same time. The strategies used to promote use of walking aids were to normalise its use by indicating to the patient how so many other elderly people in the hospital were using them, and the doctor reinforcing to him how it would prevent a further fall and fracture. Also, it was suggested that he could use an umbrella walking stick when he went out of the house. Adequate supervision by someone strong enough to catch him if he falls was required. His wife was unable to provide such supervision, hence they decided to employ a domestic helper to assist with supervision. Caregiver training was given, especially for stair climbing.

16

2) T9 compression fracture with osteoporosis confirmed on DEXA scan. Vitamin D and calcium levels were normal. As Mr TJK’s creatinine clearance was less than 35 ml/min, bisphosphonates like Alendronate are contraindicated. Although SC Denosumab given 6 monthly would be more suitable for patients with poor renal function, it was declined due to the high cost ($800+ per year). As for pain control, Paracetamol was given regularly but it was insufficient. Opioids were avoided due to previous delirium secondary to that group of analgesics. They may also inhibit the action of Madopar. NSAIDs were unsuitable due to poor renal function. IN Calcitonin was given with improvement in pain. Moreover, it has been shown to reduce new vertebral fractures (albeit in postmenopausal women). However, the incidence of adverse effects such as rhinitis, excoriations, irritation and erythema has been reported to be higher for those 75 years old and above. Mr TJK is also on CPAP for OSA which may exacerbate these adverse effects and cause discomfort, which may in turn reduce his compliance to the CPAP. Hence, regular monitoring of such signs and symptoms would be important with a view of discontinuing IN Calcitonin if they affected his CPAP administration. 3) Suboptimal diabetic control (HbA1c 7.2%). The risk of hypoglycaemia with tighter glucose control would increase an already high fall risk, and hence a HbA1c of 7.2% would be acceptable. Given that he has CKD 3 and is on Metformin, Mr TJK’s creatinine levels needed to be monitored more closely with a view of discontinuing Metformin and using other agents, e.g. insulin, to achieve glycaemic control. 4) Poorly controlled blood pressure Mr TJK’s blood pressure was initially noted to reach as high as SBP 190 mmHg while the DBP was fairly stable, ranging from 60 to 90 mmHg. There was no postural drop in blood pressure. These high SBP readings tend to coincide with back pain. Hence, once the back pain was addressed, Mr TJK’s blood pressure levels improved without needing to increase his anti-hypertensive medications. 5) Mild cognitive impairment As patients with mild cognitive impairment have an increased risk of dementia (Roberts et al 2014), it was important to encourage Mr TJK’s family to continue follow-up with the geriatrician to monitor for any progression. Moreover, continued CPAP for OSA is also important as it has been shown that the elderly are particularly susceptible to the negative

17

effects of OSA on cognition (Grigg-Damberger and Ralls 2012). His cardiovascular risk factors should also be controlled. 6) Impaired mobility and function secondary to recent fall Mr TJK underwent rehabilitation to improve his strength, balance and function. Upon discharge, he was able to ambulate with supervision using a point stick, with stairs and curbs requiring contact guard. Lower body dressing, toileting and showering also required supervision. Arrangements were made for his rehabilitation to be continued at a day rehabilitation centre. Coordinating Specialist Care Mr TJK was noted to have specialist appointments to ophthalmology for cataracts, general medicine to follow-up on the hyponatremia, geriatric medicine to follow-up on cognition, neurology for Parkinson’s disease, gastroenterology for previous gastritis and constipation, and cardiology for ischemic heart disease. Mr TJK was advised to follow-up with either a family physician or a geriatrician so as to have better coordinated care, reduced polypharmacy, and more efficient use of time and reduced transportation costs. Given that his cardiac and gastric issues are quiescent, memos suggesting discharge from those clinics to a family physician or geriatrician were given to the respective specialists. LEARNING POINTS •

Mr TJK’s case highlights the need for a holistic approach in the management of an elderly patient with a high fall risk. It also shows the importance of addressing patient’s perceptions and concerns regarding the use of walking aids.

•

This case also illustrates the deleterious effects of the “silent epidemic” of osteoporosis and the importance of screening for it so as to treat it earlier, especially for patients with a high fall risk.

•

One should be mindful of how a new treatment may impact on existing treatment (e.g. IN Calcitonin and CPAP).

•

A family physician is well placed to provide the necessary holistic care and coordinate specialist services in order to enhance patient’s care in a cost-effective manner.

18

REFERENCES Linda Resnik, Susan Allen, Deborah Isenstadt, Melanie Wasserman, Lisa Iezzoni. Perspectives on Use of Mobility Aids in a Diverse Population of Seniors: Implications for Intervention. Disabil Health J. 2009; 2(2):77–85. Roberts RO, Knopman DS, Mielke MM, Cha RH, Pankratz VS, Christianson TJ, Geda YE, Boeve BF, Ivnik RJ, Tangalos EG, Rocca WA, Petersen RC. Higher risk of progression to dementia in mild cognitive impairment cases who revert to normal. Neurology. 2014;82(4):317-25. Grigg-Damberger M, Ralls F. Cognitive dysfunction and obstructive sleep apnea: from cradle to tomb. Curr Opin Pulm Med. 2012;18(6):580-7.

19

3.1.4 Management of a Young Ventilated Patient BRIEF DESCRIPTION Mr HYS is a 28 year old Chinese gentleman who was admitted to Tan Tock Seng Hospital (TTSH) in April 2013 for fever, headache, nausea and vomiting after his return from visiting his wife’s family in a rural area in Chiang Rai. He developed acute confusion with low Glasgow Coma Scale score and was intubated. He was subsequently diagnosed to have Japanese Encephalitis which was complicated by seizures. Mr HYS also had type 2 respiratory failure possibly secondary to central apnea, and desaturations due to mucous plugging. He was started on biphasic positive airway pressure (BiPAP) and given regular suctioning. The cough assist machine was also employed to aid with expectoration of sputum. A tracheostomy was done for prolonged ventilation. Feeding was initially via nasogastric tube (NGT) and feeding pump due to oropharyngeal dysphagia characterised by decreased orolingual control and delayed swallow. However, high aspirate volumes were noted due to decreased gastric motility secondary to autonomic dysfunction. A nasojejunal tube (NJT) was subsequently inserted and patient was started on continuous feeding. This resulted in decreased aspirates. However, there were multiple episodes of NJT blockages despite conservative measures like flushing, hence requiring a number of reinsertions. His stay was further complicated by multiple episodes of nosocomial pneumonia, peripherally inserted central catheter (PICC) site infection, and sacral sores. He was subsequently transferred to the rehabilitation unit at TTSH. During his rehabilitation, Mr HYS was able to follow 2-step commands consistently and was able to focus on the task, although he was easily fatigued. He was orientated to person but inconsistent in terms of orientation to time and place. Despite 2 months of active rehabilitation, Mr HYS still required maximum assistance with forearm rollator frame for ambulation, sit to stand and transfers. He was also totally dependent for showering, toileting and feeding and required maximum assistance in grooming and dressing. He was able to verbalize with the speaking valve but it was difficult to make out what he was saying due to dysarthria. Mr HYS was referred to the Home Ventilation Team which comprised of doctors, nurses and respiratory therapists to follow-up on the use of the BiPAP and cough assist machine. After caregiver training to his wife (the primary caregiver), he was then discharged home after a 6.5 month admission. Rehabilitation was continued under the Community Rehabilitation Programme.

20

Unfortunately, Mr HYS was admitted 2 days after discharge for fever and blocked NJT. He was given a short course of antibiotics, the NJT was reinserted and he was discharged 4 days later. In view of the multiple NJT blockages, the Home Ventilation Team referred the patient to AMKH to attempt to wean off the NJT to a NGT with the secondary aim of weaning the patient off BiPAP, as it would be difficult to do both in a domiciliary setting due to the monitoring required. Management at AMKH would be shared between our team and the Consultant Family Physician from the Home Ventilation Team. On admission to AMKH, a NGT was inserted. Mr HYS’s NJT feeding regimen was progressively reduced while his NGT feeding regimen was correspondingly increased. Domperidone was initially increased to maximum dose to promote gastric emptying but he was noted by us to have QT prolongation and was hence reduced back to 10mg TDS. Erythromycin was added and gastric aspirations monitored. The NGT aspirates were reduced and the NJT was subsequently removed after successful weaning to NGT. As for weaning off BiPAP, the number of hours was initially reduced to 6 hours (10pm to 4am) from 8 hours (10pm to 6am). His arterial blood gases (ABG), symptoms of carbon dioxide retention and round the clock parameters including oxygen saturations were monitored. Initial ABG post reduction showed good oxygenation without carbon dioxide retention and BiPAP was therefore further shortened to 4 hours. However, the patient was noted to be drowsier since the reduction. Both the patient’s Home Ventilation Team Consultant and Respiratory Physician were consulted and the decision was made to increase ventilation to 6 hours. Mr HYS was less drowsy and ventilation was hence maintained at that level. The patient also had secretions which he had difficulty expectorating. Suctioning was done and the insufflator-exsufflator cough assist machine was employed to aid expectoration. Physiotherapy and occupational therapy were commenced to maintain muscle strength and also to attempt ambulation which required maximum assistance with 2 persons. Prior to discharge, certain tasks like suctioning, the use of the cough assist machine and BiPAP, NGT feeding and flushing, and assistance with transfers and ADLs were taught and/or reinforced to the patient’s wife. He was referred to the Home Nursing Foundation for NGT change. Mr HYS was then discharged home after a 1 month stay. Care was transferred back to the Home Ventilation and Community Rehabilitation Programme team who were both updated as to his progress. The option to return to AMKH for weaning BiPAP off completely and further inpatient rehabilitation in the future was also left open. To date, he has not required readmission to a tertiary hospital.

21

LEARNING POINTS •

Training and consistent professional support is crucial in the care of home ventilated patients. Hence, it is important to provide the necessary caregiver training and also to hand-off care to community partners to support caregivers.

•

The community hospital can be a good intermediate facility to facilitate specific aspects of patient care (in this case weaning off NJT and decreasing BiPAP ventilation hours). It can complement the care provided by the home care team and the tertiary hospital.

•

A shared care model of management requires open channels of communication. It is fortunate in this case that the Home Ventilation Team Consultant was able to regularly review the patient and give his input on management. The hospital’s clinical governance and other policies must facilitate such visits in order for the shared care model to succeed.

REFERENCES Evans R, Catapano MA, Brooks D, Goldstein RS, Avendano M. Family caregiver perspectives on caring for ventilator-assisted individuals at home. Can Respir J. 2012 NovDec;19(6):373-9. AK Simonds. Risk management of the home ventilator dependent patient. Thorax 2006 61: 369-371.

22

3.1.5 Management of a Patient on Continuous Ambulatory Peritoneal Dialysis BRIEF DESCRIPTION Mr TKC is an 86 year old Chinese gentleman with a background medical history of: 1) Dilated cardiomyopathy with depressed ejection fraction of 31%, cardiac resynchronization therapy device (CRT-D) inserted 2) Atrial fibrillation on Warfarin 3) End stage renal failure (ESRF), previously on haemodialysis but switched to continuous ambulatory peritoneal dialysis (CAPD) as his arteriovenous fistula collapsed. 4) Hypertension 5) Hyperlipidaemia on Ezetimibe 5mg OM and Rosuvastatin 5mg ON 6) Asthma on Seretide 25/125 2 puffs BD 7) Benign prostatic hypertrophy 8) Previous fall resulting in a right hip fracture in early 2013, internal fixation done 9) Previous peptic ulcer disease 1999 10) Sigmoid polyp (histology: tubulovillous adenoma) with an anterior resection done in June 2009 11) Previous herpes zoster He was admitted to Singapore General Hospital presenting with an altered mental state (confused speech, disorientated to place) secondary to Coagulase Negative Staphylococcus bacteraemia due to CRT-D lead infection. He was treated with IV Vancomycin for 6 weeks, followed by lifelong Bactrim to suppress any further infection. It was also noted that he had inadequate ultrafiltration volumes (possibly due to low peritoneal permeability) and he was planned for re-creation of arteriovenous fistula in 3 months’ time so as to restart haemodialysis. Mr TKC was transferred to AMK-THKH for the last dose of IV Vancomycin and rehabilitation. He was noted to have hypotension one afternoon prior to the commencement of his CAPD cycle of 80/40 (from a baseline of about 100/60). He had no giddiness and had no clinical features of a cardiac event (electrocardiogram was normal), acute blood loss or dehydration (he was above his dry weight and his intake and output balance was slightly positive). After

23

discussion with the peritoneal dialysis nurse, the dialysate used was changed from 15L of dextrose 2.5% to 10L of dextrose 2.5% and 5L of dextrose 1.5% in order to reduce the ultrafiltrate. Also, one CAPD cycle was withheld. Blood pressure was subsequently stable at his previous baseline. As Vitamin D deficiency is associated with chronic kidney disease due to the loss of 1αhydroxylase with levels inversely related to the severity of kidney disease, we screened Mr TKC for Vitamin D deficiency and he was noted to have a level of 10ug/L (deficient). Calcitriol 0.25mcg was subsequently started for him. A dual-energy x-ray absorptiometry (DEXA) scan was not done as there was no proven and safe therapy for osteoporosis for ESRF patients (the FREEDOM Trial which evaluated Denosumab did not include any patient with CKD 5). About 3 weeks into his stay, Mr TKC was noted to have fibrin strands in the peritoneal dialysis drainage bag. He had no fever or abdominal pain then. The decision was made to observe for the time being. However, a few hours later, the effluent was noted to be slightly cloudy with persistence of fibrin strands. Mr TKC also complained of some abdominal pain but remained afebrile. Vital signs were stable. He was then transferred to the Emergency Department for suspected peritonitis. LEARNING POINTS •

Although Mr TKC has ESRF, preservation of his residual renal function has beneficial effects on his mortality risk (Bargman et al). Hence, it is crucial to address complications such as hypotension which negatively impact on residual renal function.

•

The family physician can also help to screen for conditions associated with ESRF which may have been missed by the primary renal team. In this case, it was particularly important to screen for Vitamin D because a recent study by Walker et al showed that patients who are deficient in Vitamin D have worse vascular access outcomes and survival outcomes, which may explain why his arteriovenous fistula collapsed. By replacing Vitamin D, the family physician can play a part in improving the success rate of the re-creation of AVF.

•

While screening for associated conditions is part of comprehensive care, it is necessary for the family physician to think beyond just ordering the test and determine whether there is suitable therapy for the disease being screened for given each patient’s medical history.

24

•

It is important to know what the complications of peritoneal dialysis are (e.g. fibrin strands, low ultrafiltration volumes, peritonitis, etc) as there would increasingly be more patients requiring such dialysis given the rising incidence of ESRF.

•

Keeping the lines of communication open between the family physician and the primary renal team (especially the renal specialist and nurse clinician) in a shared care arrangement would facilitate the management of patients on peritoneal dialysis in the community.

REFERENCES Jamal SA, Ljunggren O, Stehman-Breen C, Cummings SR, McClung MR, Goemaere S, Ebeling PR, Franek E, Yang YC, Egbuna OI, Boonen S, Miller PD. Effects of denosumab on fracture and bone mineral density by level of kidney function. J Bone Miner Res. 2011 Aug;26(8):1829-35. Bargman JM, Thorpe KE, Churchill DN; CANUSA Peritoneal Dialysis Study Group. Relative contribution of residual renal function and peritoneal clearance to adequacy of dialysis: a reanalysis of the CANUSA study. J Am Soc Nephrol. 2001 Oct;12(10):2158-62. Walker JP, Hiramoto JS, Gasper WJ, Auyang P, Conte MS, Rapp JH, Lovett DH, Owens CD. Vitamin D deficiency is associated with mortality and adverse vascular access outcomes in patients with end-stage renal disease. J Vasc Surg. 2014 Feb 27. pii: S07415214(14)00137-2.

25

3.1.6 Role of a Community Hospital in the Care of a Palliative Patient This case was accepted and presented as a poster at the World Congress on Integrated Care 2013. BRIEF DESCRIPTION Mdm N is a 54 year old Malay lady diagnosed in 1999 with carcinoma of both breasts with metastasis to the lung, liver, axillary lymph nodes. She was subsequently noted to have cord compression due to metastasis at T8 vertebra in 2011 for which she received radiotherapy. She also underwent multiple lines of chemotherapy from 2009 to 2012, but responded poorly, developing metastasis to the cerebellar vermis which was treated with radiosurgery in July 2012. At that point in time, Mdm N stayed with her husband and two grown sons (employed) in high-rise public housing with same level access to a lift. She was able to ambulate at home with a walking frame with supervision. Community ambulation was with wheelchair with assistance. Her basic activities of daily living (bADL) were assisted by her husband (unemployed). However, she was admitted to a tertiary hospital in August 2012 for cauda equina syndrome with lower limb weakness and urinary retention secondary to bony metastasis from T7 to T10 and thoracolumbar intramedullary cord and leptomeningeal metastasis. She was started on radiotherapy to T10 to L3 and transferred to AMKH for rehabilitation. On admission, she remained independent for feeding and grooming but required assistance for dressing, toileting and showering. Her mobility was restricted to the use of the wheelchair, requiring maximal assistance for sitting to standing and transferring. With a score of 50% on the Palliative Performance Scale, her prognosis was assessed to be six months or less (Morita et al 1999). Given her poor prognosis, her rehabilitation goals as discussed with the patient were to improve her sitting tolerance in order to facilitate upper body dressing, and also to improve transferring and bed mobility. Apart from the rehabilitation element of her stay, Mdm N had nursing and medical issues which needed attention. She had a foul smelling fungating breast wound which was dressed with charcoal laced Allervyn Silver dressing in order to promote healing, reduce the risk of secondary infection, and to remove odour, which in turn improved her dignity and selfrespect. She also needed an indwelling catheter due to retention of urine. Mdm N developed neuropathic pain of both lower limbs for which Gabapentin was given with improvement. Due

26

to her immunocompromised state, Mdm N also developed zoster which resolved with Acyclovir. Unfortunately, Mdm N was noted to have an isolated left CN VI palsy. Further questioning revealed that she has had blurred vision for the past week but she did not think much of it to mention it to us. Mdm N declined computed tomography of the brain despite advice from us and her radio-oncologist that prognosis would be worse if untreated. She started having nausea which was treated symptomatically. A trial of Dexamethasone was ineffective for symptom control. With this deterioration coupled with her cauda equina syndrome, Mdm N’s function deteriorated from a modified Barthel’s Index score of 27 to 5 after six weeks of stay. Prior to this deterioration, advance care planning was initiated and Mdm N had expressed her wish to be cared for and eventually to pass away at home. She was transferred back home per her wishes after caregiver training was completed with a referral made to the Hospice Care Association for management of her nausea. However, her husband and children were unable to cope with her care on the day she returned home as she was unable to eat due to persistent vomiting and nausea. An urgent home visit was made and Mdm N was agreeable for referral to Dover Park Hospice for inpatient care. On the same day, the Dover Park home care team assessed the patient and commenced her on SC Maxolon infusion for the nausea. She was transferred to the hospice the next day and passed away 12 days later. LEARNING POINTS •

Palliative patients like Mdm N often require a multidisciplinary approach to their care. In this case, medical input was crucial in monitoring disease progression which detected the recurrence of intracranial metastasis, thereby allowing for a more precise prognostication of 2 months or less (Raymond 2001) which has a bearing on the sites of care available for this patient.

•

Pharmacologic treatment of neuropathic pain and zoster helped to improve quality of life. The physiotherapists also employed massage and transcutaneous electrical nerve stimulation which is useful for the management of neuropathic pain (Twycross 2004). Physiotherapy also helped to combat cancer-related fatigue (Narayanan and Koshy 2009).

•

The physiotherapist and occupational therapists need to set appropriate goals of rehabilitation in view of patient’s prognosis. In this case, it was to improve sitting

27

tolerance, transfers and bed mobility as these improvements would help her husband in caring for her at home. •

The medical social worker can help facilitate referrals to care providers and provide social support to families.

•

Hospices in Singapore are able to offer the necessary holistic management when families are unable to care for palliative patients at home but the patient’s prognosis needs to be 3 months or less. However, there are patients with complex care needs who have an initial prognosis of greater than 3 months such as Mdm N. This case therefore demonstrates the role a community hospital or a similar step down facility can play in the management of such palliative patients.

•

The community hospital plays a central role in organising appropriate transitions to other sites of care, with the referrals being facilitated by the medical social workers with input from the multidisciplinary team. The transitions made and the communication which occurred (in italics) to facilitate them for this case are summarised in the diagram below. Memos given for specialist appointments detailing patient’s condition

Tertiary Hospital Inpatient discharge summary

Palliative Home Care Service (linked to the hospice) Community Hospital

Inpatient discharge summary, completed ACP forms

Inpatient Hospice Home

Community hospital inpatient discharge summary, completed ACP forms, verbal communication with palliative consultant on key issues and patient’s preferences

•

It is important to address caregiver stress. In this case, the patient's husband who himself had cardiac disease was unable to assist the patient in transfers and bADLs after the diagnosis of cauda equina syndrome. Upon speaking with him, it was evident that he had caregiver stress. Hence, having the patient cared for at the community hospital brought him welcome relief from the burden of caregiving. It also allowed us to support and encourage him through this difficult period. 28

•

Admission to the community hospital presents an excellent opportunity for discussions on advance care planning to take place if it has not already been initiated. Patients are in a calmer frame of mind since the acute management of their medical issues have passed. In Mdm N’s case, her deterioration despite rehabilitation helped her to better accept her condition and prognosis. She was therefore more open to the discussion on advance care planning than when some aspects of it were broached in her first week of admission.

•

With a greater emphasis in recent years on good palliative care, clinical staff have to be adequately trained in palliative concepts of care in order to provide the holistic care necessary to promote quality of life for a patient’s last days and provide support to the family.

REFERENCES Morita T, Tsunoda J, Inoue S, Chihara S. Validity of the Palliative Performance Scale from a Survival Perspective. J Pain and Symptom Manage 1999;18:2-3. Raymond Sawaya. Considerations in the Diagnosis and Management of Brain Metastases. Oncology (Williston Park). 2001 Sep;15(9):1144-54, 1157-8. Twycross R. Factors associated with difficult-to-manage pain. Indian J Palliat Care. 2004;10:67–78. Narayanan V, Koshy C. Fatigue in cancer: A review of literature. Indian J Palliat Care. 2009;15:19–25.

29

3.1.7 Harnessing Community Services to Enable Independent Living BRIEF DESCRIPTION Mr OLG is an 83 year old Chinese ex-teacher who was premorbidly independent and community ambulant. He has a past medical history of a cholecystectomy but otherwise has no history of chronic diseases. He was conveyed to Alexandra Hospital after tripping and falling during line dancing with post-fall syncope. A CT brain showed a comminuted fracture of the right frontal bone with overlying scalp haematoma and bilateral extra-axial haematomas. A subdural haematoma (SDH) and extensive subarachnoid haemorrhage (SAH) were also noted. They were treated conservatively. Mr OLG’s stay was complicated by herpes zoster which was successfully treated with Acyclovir, benign prostatic hypertrophy (BPH) with retention of urine requiring placement of an indwelling catheter (IDC), and Enterobacter cloacae urinary tract infection likely related to IDC treated with IV Ceftriaxone. Mr OLG was also noted to have hyponatremia secondary to syndrome of inappropriate ADH secretion (SIADH) which normalised subsequently. He was transferred to AMKH for rehabilitation. Medical Problems 1) Traumatic SDH/SAH with decreased function The SAH resolved on repeat CT brain 2 months after principal event. The SDH was also reduced. Conservative management was continued by the neurosurgeons. His function plateaued at contact guard to minimal assistance for ambulation without a walking aid, stairs and curb crossing. Basic activities of daily living were generally independent except for lower body dressing and showering which required minimal assistance. 2) BPH and neurogenic bladder on IDC Mr OLG was reviewed by the urologist as he failed trial off catheter thrice. He was diagnosed with having a neurogenic bladder (cause uncertain, diabetic screen was negative) and the urologist recommended lifelong IDC and stopped the alpha-blocker. A silicon catheter was inserted prior to his discharge. 3) Longstanding bilateral hearing impairment with impacted ear wax Olive oil was instilled for one week with subsequent ear syringing done but hearing was still impaired and patient was referred to ENT for assessment and fitting of hearing aids.

30

Post Discharge Care Issues Mr OLG is single and was staying alone in a 4 room HDB flat with a lift landing on his level prior to his admission to hospital. He has 2 younger siblings who stay with their own families. He claims to have financial constraints as he allegedly draws only a small pension and has little savings. He refused to divulge further information about his financial position. He has, however, no issues with his medical bills as he is a civil service card holder. Given his need for minimal assistance for lower body dressing and showering, and contact guard to minimal assistance for ambulation, he was deemed not to be safe for independent living in the community due to the high risk of falls. He needed someone to assist him in the abovementioned activities and also for household chores (including meal preparation and grocery shopping) and travel for medical appointments should he continue living in his own home. In light of his care needs, various post discharge care options were explored with the patient and his siblings: 1) Employment of a full-time domestic helper 2) Patient shifting to one of his sibling’s residence, or one of his siblings shifting in with him 3) Stay in a voluntary welfare organisation (VWO) nursing home 4) Stay in a private nursing home However, Mr OLG’s siblings were unwilling to stay with him and vice versa. He refused to hire a domestic helper due to financial constraints and did not want to seek financial assistance from his siblings to do so. Staying at a VWO nursing home was not possible as Mr OLG refused to declare his finances. He also did not want to sell his flat to stay in a private nursing home. In any case, despite explanations of his need for assistance for ambulation and activities of daily living, Mr OLG was adamant he could take care of himself and wanted to be discharged home. Despite his intracranial haemorrhage, his AMT was 9/10 and he was assessed by the psychiatrist to have sufficient mental capacity to decide on his post-discharge care plans. The issue then was how to marshal the various community services available to facilitate independent living in his home. Firstly, Mr OLG did not own a mobile phone which would be essential should he, for example, fall down and needs to call for help. We tapped on our hospital’s patient welfare fund to purchase a mobile phone for him, with the recurrent cost of the line being paid for through his pension. Secondly, he was referred for the Singapore Programme for Integrated Care for the Elderly (SPICE) with door to door transport where he can continue to undergo

31

rehabilitation, be assisted in his showering, have his IDC changed when necessary and be provided with meals during weekdays. Meals on Wheels was arranged for the meals on weekends and public holidays. Thirdly, Ensuite Services provided by the Thye Hua Kwan Moral Society were arranged to assist with the household chores and provide medical escort as required. Mr OLG was also taught how to drain his urine bag. A home visit was conducted with the patient prior to his discharge to identify any factors which may increase his risk of falls. There were curbs at the entrance to the ensuite bathroom and kitchen, but were small enough for the patient to step over. The bathroom floor was slippery and his soap, shampoo and towel were not within easy reach. No grab bars were installed. Hence, the following recommendations were made: installation of grab bars next to the toilet bowl and sink, a seat in the shower area, a towel and amenities rack installed at the shower area, and slip resistant treatment to floor tiles. Mr OLG’s sister assisted with the necessary home modifications. Apart from the amenities and towel rack, the other modifications could be subsidised under the HDB’s Enhancement for Active Seniors (EASE) programme, requiring Mr OLG to pay only $52.50 (before GST). However, the wait for the payment process was such that it would result in too long a delay for these modifications to be completed on a timely basis and hence EASE was not pursued. Mr OLG was also encouraged to make a lasting power of attorney so as to facilitate his care should he one day lack mental capacity. He was initially keen to do so and was assessed by the psychiatrist as having the requisite mental capacity. However, after further consideration, he decided not to as he did not want to burden others with this responsibility. Mr OLG was discharged upon completion of the home modifications after a protracted length of stay of 4 months with the above support. The hospital’s transition care team was sent to ensure that he was coping sufficiently well. After about 2 weeks, his coordination of his care was handed off to the community case manager. Mr OLG was also given a discharge summary detailing his medical problems which he can present to his primary care physician.

32

LEARNING POINTS •

Elderly patients with complex care needs can be supported to live independently in their own homes if the necessary social services are provided.

•

The identification of the necessary services requires co-ordination and regular reassessment by a community case manager. The patient’s family physician is also well placed to assess the care needs and co-ordinate the services required as he or she would have a good knowledge of the patient’s medical conditions, family support and patient’s wishes. It is therefore imperative for the family physician to be kept up to date with the community services available and of any new financial assistance schemes provided by the government and welfare organisations.

•

It is often difficult to manage the discharge care issues especially for the elderly who do not have much family support. This translates to a longer length of stay and therefore cost to the government (especially since this patient is a civil service card holder). However, the avoidance of institutionalisation and greater well-being of patients when they live in their own homes also results in better quality of life and cost avoidance to the government.

•

Financial assistance initiatives such as EASE are beneficial for the elderly who live in the HDB flats. However, they can be impractical if the elderly is an inpatient and cannot be discharged until the home modifications are completed. There should be avenues where such issues can be highlighted to the relevant custodian of these financial initiatives so that tweaks can be made to support the needs on the ground.

REFERENCES EASE financial calculator: http://services2.hdb.gov.sg/webapp/DJ02FNCALV3/Main AIC portal for community services: http://www.aic.sg

33

4.2 Review Paper

Do probiotics prevent respiratory tract infections in healthy infants and children in the community? ABSTRACT Respiratory tract infections are a common cause of illness in infants and children. It results in morbidity and school absenteeism, and may consequently require caregivers to take time off work to care for the unwell child. It is also a cause of mortality especially in developing countries. A number of randomised controlled trials (RCTs) have evaluated the efficacy of probiotics in the prevention of respiratory tract infections in infants and children. PubMed and CENTRAL databases were searched for such RCTs and a total of twelve were suitable for evaluation. These RCTs involved the use of various Lactobacillus and Bifidobacterium strains in different combinations (with the majority investigating both strains) and for varying durations, although most used a treatment period of 3 or 6 months. Dosages ranged from 108 to 1010 CFU a day. The majority of studies did not show a reduction in the incidence in respiratory tract infections with the administration of prophylactic probiotics. However, probiotics may reduce the incidence and/or duration of symptoms such as fever and cough as well as the school absenteeism rate. No serious adverse events associated with probiotics administration were reported.

Keywords: Probiotics; respiratory tract infection; infants; children; preventive medicine

37

INTRODUCTION Respiratory infections such as upper and lower respiratory tract infections, sinusitis and otitis media are common causes of illness in children [1]. Apart from the associated morbidity and mortality and the consequent need for medical consultation or hospitalization, such illnesses also result in children missing school and may require parents or other caregivers to take time off to care for them, resulting in significant economic impact [2]. The substantial health and economic burden associated with respiratory infections has therefore created an impetus to discover new ways to reduce the risk of such infections. Currently, the only effective interventions apart from good hygiene are the influenza and pneumococcal vaccinations [3, 4]. However, the effectiveness of probiotic supplementation in preventing respiratory tract infections has been increasingly studied in the past decade. Probiotics are live microorganisms which confer health benefits when administered in sufficient amounts [5], and they are often included in many commercial food products such as infant and follow-on milk powder and cultured milk drinks. It is postulated that their clinical benefits are mediated through cellular and humoral immunological functions [6]. Many clinical trials have shown that certain strains can reduce the incidence of diarrhoea and also allergic diseases in children [7, 8, 9, 10]. This article aims to review the evidence on the effectiveness of probiotic supplementation in preventing respiratory illnesses in healthy infants and children in the community. METHODS Data Sources The PubMed and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched up to 19 September 2013 to identify suitable clinical trials to be reviewed. The literature search strategy used was “(respiratory tract infection) AND (children OR infants) AND (probiotics OR Lactobacillus OR Bifidobacterium OR Enterococcus faecalis)”.

38

Study Selection Criteria Randomised placebo-controlled trials (RCTs) which studied the use of probiotics in preventing respiratory illnesses in healthy children with a Jadad score of three and above were included. Studies which involved prebiotics, animal studies, those presented as abstracts in scientific conferences, those with a cross-over design and studies published in a language other than English were excluded. Selected Randomized Controlled Trials The search performed on PubMed yielded 29 potentially relevant articles after filtering for clinical trials, human studies and articles in English language. Of these 29 studies, seven had objectives not relevant to this discussion (e.g. incidence of diarrhoea as the outcome measured), seven involved the use of prebiotics, two involved non-paediatric populations, one dealt with hospitalised patients and one had a Jadad score of two. Hence, 11 were selected for review. The search performed on CENTRAL yielded 52 potentially relevant articles. Twenty-three were not RCTs, 12 were duplicated on the PubMed search, seven involved the use of prebiotics, four were not in English, four had objectives not relevant to this review, and one was not placebo-controlled, leaving only one study which was selected for review. Hence, 12 RCTs were reviewed in total from a total of 81potentially relevant articles found on PubMed and CENTRAL. The selection process is graphically represented in Figure I.

39

Figure I: Process of Selection of RCTs for Inclusion

Potentially relevant articles retrieved from PubMed applying filter for clinical trials, human studies and English language (n = 29)

Potentially relevant articles retrieved from CENTRAL (n = 52)

Articles excluded after evaluation of abstract for relevance (n = 17) • Objective of study not relevant (n= 7) • Intervention included prebiotics (n = 7) • Non-paediatric population (n = 2) • Hospitalised patients (n = 1)

Articles to be assessed for quality (n = 12)

Articles excluded (n = 51) • Not RCT (n = 23) • Duplicated on PubMed search (n = 12) • Intervention included prebiotics (n = 7) • Studies not in English (n = 4) • Objective of study not relevant (n = 4) • No placebo (n = 1)

Articles excluded after evaluation of quality (n = 1) • Jadad score 2

1 RCT from CENTRAL qualifying for inclusion

11 RCTs from PubMed qualifying for inclusion

Total of 12 RCTs selected for inclusion

40

Data Extraction The study design, Jadad score, characteristics of the study population, type, dosage, duration and method of administration of probiotics and placebo, primary and secondary outcomes, results and adverse effects were extracted from the included RCTs. They are summarised in Table I.

41

Table I: Summary of Data Extracted from Selected RCTs Reference

Jadad Score

Study Population

Kumpu et al (2012) [11]

4

523 children aged 2 – 6 years old in childcare centres in Finland

Type of probiotic / placebo and form of administration Lactobacillus rhamnosus GG in milk vs same milk without probiotics

Duration / dose of treatment

Outcome of Interest

Results

Adverse Events (AE)

Remarks

3 times a day for 5 days a week over a period of 28 days (average daily milk consumption was 400mL in both groups, equivalent 8 to ~ 10 CFU of probiotics per day in the GG group)

Number of days with respiratory symptoms, number and duration of respiratory symptom episodes, number of diagnosed respiratory tract infections and number of antibiotic treatments.

No difference in the number of days with respiratory symptoms (5.03/month vs 5.17/month, p=0.098), median duration of respiratory symptoms, number of diagnosed respiratory tract infections and number of antibiotic treatments.

8 in probiotics group and 14 in placebo group.

Lactobacillus rhamnosus GG recovered in the faecal sample of 31% of the children in the placebo group, possibly due to widespread availability of GG containing products in Finland.

Reduction in the average number of days with at least 1 respiratory symptom (4.71/month vs 5.67/month, IRR 0.83, p<0.001), only after subanalysis of completed cases (n=81 in probiotics group

42

15 (probiotics group = 5, placebo = 10) had GI problems (nausea, abdominal pain) 7 (probiotics group =3, placebo =4) skin problems e.g. rash. No serious AE.

Run in period of 2-3 weeks too short to clear the GG as 34% of subjects still carried GG at the beginning of the intervention. RTIs were diagnosed by healthcare professionals. Reduction in the number of days with at least 1 respiratory symptom of less

and n=47 for placebo). Rerksuppaphol and Rerksuppaphol (2012) [12]

5

80 schoolchildren aged 8-13 years old from a public school in rural Thailand.

Lactobacillus acidophilus 9 (minimum 10 CFU/capsule) + Bifidobacterium bifidum (minimum 9 10 CFU/capsule) Placebo: oral rehydration salts powder in identical-coloured capsules.

Twice a day for 3 months

Occurrence of any symptom of cold. Use of antibiotics, school absence due to any cause, school absence due to cold, duration of all symptoms.

Reduction in occurrence of any symptom of cold (77% vs 95%, p=0.048). Reduction in OR for fever 0.32 (95% CI 0.12 – 0.81), cough 0.17 (95% CI 0.05 – 0.58), rhinorrhoea 0.18 (95% CI 0.04 – 0.79), cold-related school absence 0.10 (95% CI 0.02 – 0.50), after adjustment for age, sex and BMI. Shorter median duration of all symptoms (fever 0 vs 1 days, p=0.010, cough 2 vs 7 days, p=0.035, rhinorrhoea 3.5 vs 6.5 days, p=0.044). Reduction in

43

1 in probiotics group had fever and vomiting.

than 1 day is not clinically significant. Unknown whether the different strains of probiotics have different efficacies (may have synergy or the added beneficial effect may simply be due to a higher probiotic dose).

school absence due to any cause (8% vs 45%, p=0.000) and due to cold (5% vs 35%, p=0.001). No difference in the use of antibiotics (5% vs 13%, p=0.432). Agustina et al (2012) [13]

5

494 healthy children 1 – 6 years old, from low socioeconomic communities, not breastfed in East Jakarta, Indonesia.

4 groups: Low calcium milk (LC), regular calcium milk (RC),RC with Lactobacillus casei CRL431 8 5x10 CFU, RC with Lactobacillus reuteri 8 DSM17938 5x10 CFU.

180ml of milk twice daily for 6 months

Number and duration of acute respiratory tract infection (ARTI) episodes

Probiotics contained within straws used to drink the milk.

No significant difference in the number of ARTI episodes and duration of episodes.

Nil

ARTI defined as ≥ 1 respiratory tract symptoms (runny nose, cough, sore throat) and /or ≥1 additional respiratory tract symptom or 1 constitutional symptom (fever, headache, restlessness, aphony, SOB, acute ear pain or discharge), confirmed by physician. Duration defined as number of consecutive days with ≥2 defined signs and symptoms with a 7 day symptom

44

free interval before it is considered a new episode. Taipale et al (2011) [14]

5

109 1 month old Finnish infants

Bifidobacterium animalis subsp lactis (BB-12) 9 5x10 in a xylitol tablet vs control xylitol tablet.

Twice a day (total BB-12 9 10 CFU a day) for 6 – 7 months via slow release pacifier or spoon.

Incidence of ARTI and doctordiagnosed AOM.

Reduction in the incidence of ARTI (65% vs 94%, RR = 0.69, p=0.014).

Nil

Only 69 completed the study. ARTI defined as at least 2 infectious symptoms (runny nose, nasal congestion, cough, shortness of breath) over one day, or 1 symptom over 2 consecutive days.

No difference in doctordiagnosed AOM.

Faecal sample collected for BB12 determination – 62% in treatment arm had it isolated and 17% of controls. Xylitol may be considered a prebiotic, but levels consumed not significant as a prebiotic. Merenstein et al (2010) [15]

5

638 3 - 6 year old children in Washington

Fermented dairy drink with Lactobacillus

90 consecutive days

Incidence of URTI and LRTI.

45

Reduced incidence of URTI by 18%

18 subjects had at least 1 AE vs 22 in

Yoghurt starters just improve lactose digestion

DC metropolitan area in schools/day care.

Hojsak et al (2009) [16]

5

281 children attending day care centres in Zagreb, Croatia (age range from 13 to 86 months).

casei DN-114 001 8 (1x10 CFU/g) with yoghurt starters (Streptococcus thermophilus and Lactobacillus 7 bulgarius >10 CFU/g) vs nonfermented dairy drink in 200g bottles.

Lactobacillus rhamnosus strain 9 GG 10 CFU in 100ml of fermented milk product vs post – pasteurized fermented milk product without LGG

Change of behaviour because of illness (any illness that resulted in a change in activity such as missed school, birthday party, soccer game, etc).

Daily for 3 months

Number of children with RTI. Number of children with RTI > 3 days. Total number of days with respiratory and GI symptoms. Number of days absent from day care

46

(0.027 vs 0.033, IR 0.82 p=0.036). No significant difference in incidence of LRTI, rate of cough, runny nose, fever.

Fewer number of children with URTI (41.7% vs 66.9%, RR 0.66 p <0.001, NNT 5). Fewer number of RTI longer than 3 days (28.1% vs 49.3%, p <0.001, RR 0.57 NNT 5). Reduction in absence from

control group, 1 serious AE vs 2 in control group. All participants with serious AEs were hospitalised with spontaneous resolution and believed not to be related to study product.

Nil

and do not survive intestinal transit in significant quantities to have a positive impact on intestinal health. Infections were categorized by research personnel based on parentreported symptoms. No mention of the diagnostic criteria used by the research personnel. Episodes of RTI diagnosed by doctors, but no strict criteria for diagnosis.

Leyer et al (2009) [17]

4

326 healthy children 3 to 5 years old from Zhejiang Province, China.

1g sachet of Lactobacillus acidophilus HCFM, or Lactobacillus acidophilus NCFM and Bifidobacterium Animalis subsp lactis Bi-07, with sucrose as diluent, vs sucrose alone mixed in 120ml of 1% fat milk

2 times a day x 7days a week for 6 months (total 1 x 10 10 CFU a day of probiotics in intervention groups)

due to infections.

day care centre due to GITI or RTI (3.1 vs 5.1 days, p<0.001).

Frequency and duration of fever, cough and rhinorrhoea.

Adjusted for age and time absent from child care:

Incidence of physicians’ visits and antibiotic prescription. Number of days absent from school.

Lactobacillus acidophilus vs placebo: Reduction in frequency of fever (OR 0.57), cough (OR 0.59), antibiotics (OR 0.35), p all < 0.03. Lactobacillus acidophilus + Bifidobacterium lactis vs placebo: Reduction in frequency of fever (OR 0.34), cough (OR 0.44), rhinorrhoea (OR 0.52), antibiotics (OR 0.23). Shorter symptom duration than

47

Nil

Parent administered questionnaire with no physician confirmation of infection. Study indicates a trend toward more significant results with combination vs single-strain preparations which may be due to bifidobacteria in the mouth which decreases adherence of certain respiratory viruses. However, no mucosal swabs were done in this study to confirm this.

placebo for both Lactobacillus acidophilus + Bifidobacterium lactis group (3.2 ± 0.76 days, p<0.001) and Lactobacillus acidophilus group (-2.17 ± 0.71, p 0.0023). Fewer number of days absent for both probiotic groups compared to placebo (Lactobacillus acidophilus group: -1.6 days, p=0.01, Lactobacillus acidophilus + Bifidobacterium lactis group: 1.4 days, p=0.01). No significant differences in outcomes between probiotic groups. Rautava et al (2009) [18]

5

81 Finnish infants 2 months old at

Lactobacillus rhamnosus GG and

10

1x10 CFU daily for 10 months

Incidence of RTI and doctor-

48

Reduction in the incidence of AOM (RR 0.44,

Nil

Diagnostic criteria given for AOM, but not for

time of entry

diagnosed AOM.

Bifidobacterium lactis Bb-12 vs placebo mixed in infant formula.

Incidence of recurrent (≥ 3) infections, need for antibiotics or tympanostomy st during 1 year of life.

respiratory infections.

p=0.014) and antibiotic use (RR 0.52, p=0.015). Reduction in the incidence of recurrent infections (28% vs 55%, RR 0.51, p=0.022). No significant difference in incidence of RTI in the first 7 months of life, or in the incidence of tympanostomy in the first year of life.

Hatakka et al (2007) [19]

5

309 otitis prone Finnish children (10 months to 6 years).

Lactobacillus rhamnosus GG, Lactobacillus rhamnosus LC, Bifidobacterium breve, Propionibacterium freudenreichii ssp shermanii JS (8-9 9 x 10 CFU/capsule of each strain) vs cellulose microcrystalline capsule.

1 capsule daily for 6 months

Occurrence and duration of AOM episodes. Frequency of pathogen carriage. Occurrence of URTI episodes. Number of antibiotic treatments.

49

No difference in the number and duration of AOM episodes, URTI episodes, or antibiotic treatment. Nasopharyngeal carriage of potential otitis pathogens not reduced. Reduction in the number of participants with

Not mentioned

URTI was defined while AOM was diagnosed by a senior resident in ENT.

≥ 4 URTI episodes (OR 0.56, 95% CI 0.31-0.99, p=0.046), but not statistically significant when looking at ≥ 2, 3, 5, or 6 episodes. Wiezman et al (2013) [20]

3

201 healthy Israeli 4 to 10 months old attending child care.

Humanized cow’s milk formula with Bifidobacterium lactis BB-12 7 (1x10 CFU/g), or with Lactobacillus 7 reuteri (1x10 CFU/g), or with no probiotics.

12 weeks

Number of episodes and days with fever. Number of episodes and days with respiratory symptoms. Number of clinic visits, antibiotic prescriptions and child care centre absences.

The Lactobacillus reuteri group had statistically significant fewer days with fever (0.17 vs 0.83 and 0.86 respectively), clinic visits (0.23 vs 0.55 and 0.51), absences from child care (0.14 vs 0.43 and 0.41), antibiotic prescriptions (0.06 vs 0.19 and 0.21) than the control or BB-12 group respectively. Rate and duration of respiratory illnesses did not differ

50

Nil

No mention of how much milk powder was prescribed, but mean daily ingested dose of probiotics was reported as 1.2x 9 10 CFU/day. Pediatrician diagnosed illnesses. Results not clinically significant (e.g. difference in duration of fever < 1 day).

significantly between groups. Hatakka et al (2001) [21]

5

594 Finnish children 1 to 6 years old attending day care in Helsinki.

1% fat milk with Lactobacillus rhamnosus GG 55 10x10 CFU/ml vs placebo without probiotics.

3 times a day, 5 days a week for 7 months (mean of 260ml taken per day, 18 2x10 CFU of probiotics a day)

Number of days with respiratory symptoms. Absences from day care centre due to illness. Number of children with AOM, sinusitis, acute bronchitis and pneumonia. Number of children given antibiotic treatment for respiratory infections.

51

No significant differences between the groups in all of the outcomes listed after adjustment for age. Unadjusted results revealed fewer days of absence from day care (4.9 vs 5.8, p=0.03), and an absolute reduction in antibiotic treatment of 9.6% (p=-0.03) in the probiotics group.

Nil

There were more 3 and 5 year olds in the control group and more 4 and 6 year olds in the Lactobacillus group, hence age adjustment was applied which reduced the significance of some results. Complications were diagnosed by a doctor. Faecal samples (randomly selected) taken at beginning, middle and end of study. 15% of the children in the control group had Lactobacillus GGlike bacteria isolated from their stool samples. They may have unwittingly consumed it during the study.

Smerud et al (2007) [22]

4

240 children aged 12 to 36 months attending day care in Oslo, Norway.

Milk drink containing Lactobacillus rhamnosus GG, Lactobacillus acidophilus LA-5 8 (10 CFU/ml) and Bifidobacterium 7 BB-12 (10 CFU/ml) vs fermented milk drink without probiotics.

1.5dL daily for 7 months

Number of days with symptoms of respiratory infection.

No significant difference in the number of days with respiratory symptoms (25.4 vs 25.1 days, p=0.63).

Not mentioned

Respiratory symptoms defined o as fever ≥ 37.5 C, cough, sore throat, runny nose, ear pain and wheezing. Respiratory infection defined as having 2 or more symptoms. Only 10 out of 97 in the probiotics group and 9 out of 102 in the control group were per protocol, thus limiting the validity of the study. 169 participants were not per protocol as they were deemed not compliant to milk intake (defined as intake < 60% on days away from day care). 75 had missing observations of symptoms for > 20 days.

Abbreviations: AOM: acute otitis media, IR: incidence risk, RTI: respiratory tract infections, CI: confidence interval, LRTI: lower respiratory tract infections, URTI: upper respiratory tract infections, GI: gastrointestinal , OR: odds ratio, GITI: gastrointestinal tract infections, RR: relative risk

52

RESULTS Characteristics of Study Design and Jadad Scores All the selected RCTs were randomised, double-blinded, placebo controlled trials. Eight of them had a Jadad score of five [12, 13, 14, 15, 16, 18, 19, 21], three had a score of four [11, 17, 22], and one had a score of three [20]. Characteristics of the Study Population The study populations of the included RCTs range from 1 month old to 13 years old. Three RCTs involved infants less than 1 year old [14, 18, 20], while eight RCTs targeted toddlers and those in the pre-school years [11, 13, 15, 16, 17, 19, 21, 22]. Only one RCT involved mainly primary school aged children [12]. Four RCTs were conducted in developing countries [12, 13, 16, 17]. One study targeted children prone to otitis media [19]. Probiotic Interventions The Lactobacillus and Bifidobacterium strains were primarily used as interventions in the RCTs. Eight RCTs investigated the use of both strains [12, 13, 15, 17, 18, 19, 20, 22], three RCTs explored the use of only Lactobacillus strains [11, 16, 21], while only one RCT investigated only the Bifidobacterium strain [14]. The dosage of probiotics ranged from 108 to 1010 CFU per day. The duration of treatment varied from 28 days to 10 months, with most studies using a treatment period of 3 or 6 months. The form of administration was mainly admixed in milk for ease of administration, with one study involving infants using a slowrelease tablet delivered via pacifier or spoon [14], and another study involving 8 to 13 year old children using a capsule [12]. Compliance to the interventions was mainly through records kept by the person administering the interventions, namely either teachers or parents. Only three studies evaluated the amount of probiotics isolated from stool samples taken from participants [11, 14, 21].

53

Outcomes The main outcomes in the selected RCTs can be broadly categorised into those relating to respiratory tract infections (e.g. number of episodes, duration of episodes) and those relating to respiratory symptoms (e.g. number of participants with fever, runny nose, cough, etc, duration of symptoms). Most studies relied on parents to record such infections or symptoms, though some required physician diagnosis. This review will also look at whether there are any differences in absenteeism rates as this has an impact on whether caregivers need to take time off work to care for their children. Outcomes Regarding Respiratory Tract Infections Eight of the 12 RCTs analysed outcomes relating to respiratory tract infections [11, 13, 14, 15, 16, 18, 19, 21], with most being conducted in developed countries [11, 14, 15, 18, 19, 21]. Of these eight studies, only three studies found a statistically significant reduction in the incidence of respiratory infections in the probiotic group with a relative risk ranging between 0.66 and 0.82 [14 – 16]. Two studies used a daily dose of probiotics of 109 CFU [14, 16], while one used 108 CFU [15]. This is not too dissimilar from the five negative studies where three studies used 108 CFU [11, 13, 21], one used 109 CFU [19], and one used 1010 [18]. However, only one positive study stated the criteria used to diagnose respiratory tract infections (but not doctor-diagnosed) [14], while for four of the five negative studies, respiratory tract infections were either doctor-diagnosed and/or the diagnostic criteria were stated [11, 13, 19, 21]. There were no significant differences in the economic status of the participants for both the positive and negative studies. With regards to the incidence of acute otitis media, one Finnish study involving 81 infants found that while the incidence of respiratory infections was not reduced, the incidence of recurrent respiratory infections (defined as three or more episodes) as well as the incidence of acute otitis media (doctor-diagnosed based on a set criteria) were significantly reduced with a relative risk of 0.51 and 0.44 reported respectively [18]. However, another small

54

Finnish study involving infants reported a reduction in the episodes of respiratory infections (relative risk of 0.69; p = 0.014) but no difference in doctor-diagnosed acute otits media [14]. Two large studies also did not find a difference in doctor-diagnosed acute otitis media [19, 21]. Outcomes Regarding Respiratory Symptoms Seven of the 12 RCTs had outcomes relating to respiratory symptoms [11, 12, 15, 17, 20, 21, 22]. Four of these studies had positive results for the probiotic group [11, 12, 17, 20]. The majority of these studies used a daily probiotic dose of greater than or equal to 108 CFU [11, 12, 17], while all the negative studies used a daily dose of 108 CFU. There were no significant differences in treatment durations. All three of the negative studies were conducted in developed countries, while two of the four positive studies were conducted in developing countries [12, 17]. Among the four positive studies, one study in older children found a statistically significant reduction in occurrence of fever (odds ratio (OR) 0.32), rhinorrhoea (OR 0.18) and cough (OR 0.17) [12], while a study involving infants found a reduction in the number of days with fever (0.83 in the placebo group vs 0.86 in the Bifidobacterium group and 0.17 in the Lactobacillus group; statistically significant) [20]. A study in pre-school children found a reduction in the incidence of fever (OR 0.34), cough (OR 0.44) and rhinorrhoea (OR 0.52) with the use of combined probiotic strains (Lactobacillus and Bifidobacterium ) as compared to placebo [17]. Another study in that age group found a reduction in the number of days with at least one respiratory symptom (4.71 per month vs 5.67 per month in the placebo group) but only on subanalysis of completed cases which was defined as participants in the intervention group demonstrating new probiotic colonisation of the gastrointestinal tract and participants in the placebo group showing colonisation below detection limits before and after the intervention period [11]. However, this reduction of less than 1 day is not clinically significant.

55

Outcomes Regarding Absence from School or Child Care Six of the 12 RCTs explored whether probiotics had any effect on the number of days absent from school or day care, with four studies showing benefit with probiotics [12, 16, 17, 20]. One study involving a combined Lactobacillus and Bifidobacterium intervention reported a reduction in cold-related school absence in the combined probiotic group (OR 0.10) [12]. Another study also involving combined probiotics showed a reduction in the average number of days absent from child care due to any cause of 1.4 days as compared to placebo [17]. However, that same study showed a slightly greater reduction of 1.6 days in the Lactobacillus only group compared to placebo, although there was no significant difference when this group was compared to the combined probiotic group [17]. Another study involving only the use of Lactobacillus showed a lower number of missed day care days due to respiratory or gastrointestinal tract infections of 3.1 days vs 5.1 days for the placebo group (p < 0.001) [16]. A study involving infants showed a mean of 0.14 days of absence from child care due to any cause for the Lactobacillus group (vs 0.43 for the placebo and 0.41 for the Bifidobacterium group), which is not clinically significant [20]. However, two studies involving Lactobacillus did not demonstrate any advantage in the probiotic group in reducing absenteeism. One showed no significant differences in the number of days of absence from day care after adjustment for age [21], while the other did not demonstrate any change of behaviour (e.g. missed school, birthday party, soccer game, etc) due to respiratory or gastrointestinal tract infections [5]. Adverse Events No serious adverse events attributable to the probiotic study products were reported in 10 RCTs [11, 12, 13, 14, 15, 16, 17, 18, 20, 21], while the other two did not report data relating

56

to adverse events. Three studies reported minor problems like nausea, vomiting, constipation, flatulence, abdominal pain and rash [11, 12, 15]. DISCUSSION The efficacy of probiotics for the prevention of illness has been widely investigated. In particular, its health benefits in preventing respiratory tract infections in children have gained much interest in the past decade as commercial products such as milk powder are increasingly being supplemented with probiotics. These bacteria have been shown to induce immunomodulatory activity including the recruitment of CD4+ T-helper cells [23], and affect antigen-specific IgG1/IgG2 antibodies [24]. In the prevention of acute otitis media, it is postulated that both the reduction of colonisation by pathogens such as Streptococcus pneumoniae by local inhibition and immunomodulation throughout the common mucosa immune system [18]. This review shows that the administration of prophylactic probiotics for infants and children does not reduce the incidence of respiratory tract infections as demonstrated by the majority of the selected RCTs. Probiotics may reduce the incidence and/or duration of symptoms such as fever and cough, and the absenteeism rate from school or child care. However, fever may be due to causes other than respiratory tract infections, and the reduction in duration of symptoms may not be clinically significant. Also, most of the studies showing reduced absenteeism did not look specifically at cold-related absenteeism and the result for one study was not clinically significant. The majority of RCTs did not report any serious adverse events associated with probiotics administration with a few studies reporting only minor problems. The administration of more than one strain of probiotics seem to also result in a reduction in respiratory infections and/or symptoms [12, 15, 17, 18], possibly due to the resultant higher total dose of probiotics administrated, although two RCTs show no difference when compared to placebo [19, 22],

57

with the postulation that when multi strains of probiotics are used, one strain may inhibit the effect of other strains [15, 19]. Studies involving both prebiotics and probiotics (i.e. synbiotics) were excluded in order to preclude the possible confounding effect of prebiotic administration. It is, however, worth noting that some studies involving synbiotics show a reduction in the incidence of upper respiratory tract infections [26], as well as pneumonia and severe acute lower respiratory tract infections [27]. It is also interesting to note that a study conducted on hospitalised children showed the beneficial effects of probiotics in reducing the risk of nosocomial respiratory tract infections [28]. Several limitations should be considered in the interpretation of the findings of this review. Firstly, the study populations are diverse geographically. It has been mooted that lack of positive results in the probiotics groups may be due to the high prevalence of commercially available probiotic-containing products which may be inadvertently consumed by the placebo groups for studies conducted in developed countries. Some studies assessed the extent of inadvertent consumption, with Kumpu et al (2012) finding that 31% of the participants in the control group had Lactobacillus isolated from their stool samples [11], while other Finnish studies by Taipale et al (2011) and Hatakka et al (2001) reported a rate of 17% and 15% respectively [14, 21]. Secondly, there is also a wide range of duration of probiotic treatment. Thirdly, the evaluated outcomes were also heterogeneous. Moreover, different studies had different definitions of what constituted a respiratory tract infection, hence making comparisons between studies difficult. In addition, these infections were parent-reported (and presumably less reliable) in some studies, while physician diagnosed in others. This could explain why most of the studies exploring the incidence of respiratory tract infections did not find any difference as most of the negative studies (four of the five) relied on a doctor’s diagnosis and/or the use of diagnostic criteria, while only one of the three positive studies mentioned the use of

58

diagnostic criteria. This heterogeneity in outcomes and definitions also makes a metaanalysis on this topic challenging. Finally, only the PubMed and CENTRAL databases were searched, and RCTs other than in English were excluded. CONCLUSION The majority of RCTs reviewed indicate that prophylactic probiotics do not reduce the incidence of respiratory tract infections in healthy infants and children, although probiotics may reduce the incidence and duration of symptoms of respiratory illness. Although probiotic administration is generally safe with only mild adverse events being reported, the evidence does not clearly demonstrate a definite benefit in the administration of probiotics to healthy infants and children in terms of a reduction in respiratory infections and symptoms and coldrelated school absenteeism rates.

59

REFERENCES 1. Meissner HC. Reducing the impact of viral respiratory infections in children. Pediatr Clin North Am. 2005 Jun;52(3):695-710. 2. Yin JK, Salkeld G, Lambert SB, Dierig A, Heron L, Leask J, Yui Kwan Chow M, Booy R. Estimates and determinants of economic impacts from influenza-like illnesses caused by respiratory viruses in Australian children attending childcare: a cohort study. Influenza Other Respir Viruses. 2013 Nov;7(6):1103-12. 3. Katz JA, Capua T, Bocchini JA Jr. Update on child and adolescent immunizations: selected review of US recommendations and literature. Curr Opin Pediatr. 2012 Jun;24(3):407-21. 4. Lee BY, Shah M. Prevention of influenza in healthy children. Expert Rev Anti Infect Ther. 2012 Oct;10(10):1139-52. 5. Borchers AT, Selmi C, Meyers FJ, Keen CL, Gershwin ME. Probiotics and immunity. J Gastroenterol. 2009;44(1):26-46. 6. Boirivant M, Strober W. The mechanism of action of probiotics. Curr Opin Gastroenterol. 2007 Nov;23(6):679-92. 7. Guandalini S. Probiotics for prevention and treatment of diarrhoea. J Clin Gastroenterol. 2011 Nov;45 Suppl:S149-53. 8. Reid G, Jass J, Sebulsky MT, McCormick JK. Potential uses of probiotics in clinical practice. Clin Microbiol Rev. 2003 Oct;16(4):658-72. 9. Sanders ME, Guarner F, Guerrant R, Holt PR, Quigley EM, Sartor RB, Sherman PM, Mayer EA. An update on the use and investigation of probiotics in health and disease. Gut. 2013 May;62(5):787-96.

60

10. Kuitunen M. Probiotics and prebiotics in preventing food allergy and eczema. Curr Opin Allergy Clin Immunol. 2013 Jun;13(3):280-6. 11. Kumpu M, Kekkonen RA, Kautiainen H, et al. Milk containing probiotic Lactobacillus rhamnosus GG and respiratory illness in children: a randomized, double-blind, placebocontrolled trial. Eur J Clin Nutr. 2012 Sep;66(9):1020-3. 12. Rerksuppaphol S, Rerksuppaphol L. Randomized controlled trial of probiotics to reduce common cold in schoolchildren. Pediatr Int. 2012 Oct;54(5):682-7. 13. Agustina R, Kok FJ, van de Rest O, et al. Randomized trial of probiotics and calcium on diarrhoea and respiratory tract infections in Indonesian children. Pediatrics. 2012 May;129(5):e1155-64. 14. Taipale T, Pienihäkkinen K, Isolauri E, et al. Bifidobacterium animalis subsp. lactis BB-12 in reducing the risk of infections in infancy. Br J Nutr. 2011 Feb;105(3):409-16. 15. Merenstein D, Murphy M, Fokar A, et al. Use of a fermented dairy probiotic drink containing Lactobacillus casei (DN-114 001) to decrease the rate of illness in kids: the DRINK study. A patient-oriented, double-blind, cluster-randomized, placebo-controlled, clinical trial. Eur J Clin Nutr. 2010 Jul;64(7):669-77. 16. Hojsak I, Snovak N, Abdović S, Szajewska H, Misak Z, Kolacek S. Lactobacillus GG in the prevention of gastrointestinal and respiratory tract infections in children who attend day care centers: a randomized, double-blind, placebo-controlled trial. Clin Nutr. 2010 Jun;29(3):312-6. 17. Leyer GJ, Li S, Mubasher ME, Reifer C, Ouwehand AC. Probiotic effects on cold and influenza-like symptom incidence and duration in children. Pediatrics. 2009 Aug;124(2):e172-9.

61

18. Rautava S, Salminen S, Isolauri E. Specific probiotics in reducing the risk of acute infections in infancy—a randomised, double-blind, placebo-controlled study. Br J Nutr. 2009 Jun;101(11):1722-6. 19. Hatakka K, Blomgren K, Pohjavuori S, et al. Treatment of acute otitis media with probiotics in otitis-prone children-a double-blind, placebo-controlled randomised study. Clin Nutr. 2007 Jun;26(3):314-21. 20. Weizman Z, Asli G, Alsheikh A. Effect of a probiotic infant formula on infections in child care centers: comparison of two probiotic agents. Pediatrics. 2005 Jan;115(1):5-9. 21. Hatakka K, Savilahti E, Pönkä A, et al. Effect of long term consumption of probiotic milk on infections in children attending day care centres: double blind, randomised trial. BMJ. 2001 Jun 2;322(7298):1327. 22. Smerud HK, Kleiveland CR, Mosland AR, Grave G, Birkeland S. Effect of a probiotic milk product on gastrointestinal and respiratory infections in children attending day-care. Microb Ecol Health Dis. 2008;20(2):80-85. 23. Connolly E, Valeur N, Engel P, Carbajal N, Ladefoged K. In situ colonization and immunomodulation by the probiotic Lactobacillus reuteri in the human gastrointestinal tract. Clin Nutr 2003;22:S57 24. Maassen CB, Boersma WJ, van Holten-Neelen C, Claassen E, Laman JD. Growth phase of orally administered Lactobacillus strains differentially affects IgG1/IgG2 ratio for soluble antigens: implications for vaccine development. Vaccine 2003;21(21-22):2751-7. 25. Timmerman HM, Joning CJ, Mulder L, et al. Monstrain, multistrain and multispecies probiotics – a comparison of functionality and efficacy. Int J Food Microbiol 2004; 96(3): 21933.

62

26. Maldonado J, Cañabate F, Sempere L, et al. Human milk probiotic Lactobacillus fermentum CECT5716 reduces the incidence of gastrointestinal and upper respiratory tract infections in infants. J Pediatr Gastroenterol Nutr. 2012 Jan;54(1):55-61. 27. Sazawal S, Dhingra U, Hiremath G, et al. Prebiotic and probiotic fortified milk in prevention of morbidities among children: community-based, randomized, double-blind, controlled trial. PLoS One. 2010 Aug 13;5(8):e12164. 28. Hojsak I, Abdović S, Szajewska H, Milosević M, Krznarić Z, Kolacek S. Lactobacillus GG in the prevention of nosocomial gastrointestinal and respiratory tract infections. Pediatrics. 2010 May;125(5):e1171-7.

63