By Dr. Juliet Otieno

This the second blog in a series from the Pneumonia Etiology Research for Child Health or PERCH study team. PERCH is a multi-country case-control study of the etiology of severe and very severe pneumonia in children aged 1-59 months. Led by IVAC and funded by the Bill & Melinda Gates Foundation, PERCH operates seven research sites throughout Africa and Asia.

As a Kenyan doctor, having completed my internship at a rural mission hospital and worked as a junior doctor in a paediatric hospital for just over a year, I found myself two years ago in a small rural town, Kilifi, in the heart of a study on the causes of childhood pneumonia, the Pneumonia Etiology Research for Child Health or PERCH. When I began working with the PERCH study, a study on the “causes” of pneumonia initially seemed to me as a rather counter-intuitive approach to the problem at hand. I had spent years cramming the national diagnostic and treatment guidelines for pneumonia and I was pretty confident in the management of children with acute and chronic respiratory illnesses. I did not need the much-cited research to know what pneumonia looks like or to understand its burden on our children.

physicians examine chest x-ray

Physicians from three different PERCH sites, Dr. Juliet Otieno, Dr. Bernard Ebruke, and Dr. David Moore, examine a chest x-ray of a patient with pneumonia during a cross-site visit in South Africa.

Childhood pneumonia is a topic I have always been able to easily relate to, starting from my days as a medical student in the busy paediatric wards where most of the very sick children had respiratory illnesses. While efforts have been instituted to ensure early recognition and treatment of children with pneumonia, unfortunately a child succumbing to death from a respiratory illness is still seen as an unfortunate but inevitable eventuality in most of our hospitals in Kenya.

Two years later, I have had many epiphanies along the way about the importance of investigating the causes of pneumonia, but a recent one has been particularly influential and has changed my mindset completely. I recently had the opportunity to attend a cross-site visit to Chris Hani Baragwanath Hospital in Johannesburg, South Africa – another PERCH study site – where I had a chance to look at a histopathology slide of lung tissue from a child who died of pneumonia. Post-mortem lung biopsy studies are carried out as part of the PERCH study in some of the participating sites because they can give researchers a much more direct, accurate diagnosis of etiology in fatal pneumonia cases. (Learn more in this 2011 article in Clinical Infectious Diseases.)

This particular post-mortem lung specimen we were examining was from a child who presented with clinical features consistent with pneumonia caused by pulmonary tuberculosis and was treated for this condition without success and died. The post-mortem analysis revealed that the child did not have TB, a bacterial infection, but instead the pneumonia had been caused by florid cytomegalovirus infection, a virus that spreads through bodily fluids and is especially dangerous for patients with compromised immune systems. If the doctors caring for him had known this, they would have treated him differently, and he might have lived. Unfortunately, we cannot get such an accurate picture of the pathogens present in the lung with current tests available for patients. It dawned on me just then how important post-mortem results are in defining the epidemiology of pneumonia in a particular setting, made possible through studies like PERCH, and which help doctors see the range of pathogens that are causing child deaths in their hospitals, so they might be more aware and ready to treat them.

I left wondering what definitive pathogens we would find if we carried out post-mortem studies on fatal pneumonia cases in Kilifi, and how this would influence our clinical management of pneumonia. Would we only find the “traditional” bacteria we expect? Or would we be opened to a whole new world of “atypical” bacteria that we occasionally treat as a last ditch resort? Would we see that the recently introduced vaccine against the streptococcal pneumonia bacteria caused a significant reduction in one of the most common “traditional” pneumonia-causing bacteria? Would our very definition of “atypical” pathogens be challenged? Would we be more confident in administering relevant antimicrobial treatment – albeit a non-conventional approach – in a more timely manner? I also wondered how post-mortem findings would compare to the induced sputum test results we gather through PERCH, which get us as close as we can to the lung while the child is living, but are still unproven, and leave us wondering if we’re really getting at the source of the infection. Most importantly, I wondered if all this knowledge could be the answer needed to save lives, reduce unnecessary use of antibiotics that can lead to resistance, and decrease the burden of pneumonia morbidity in our paediatric ward?

Otieno demonstrates NP swab

Dr. Otieno demonstrates a nasopharyngeal swab during a training session.

At a clinical level, results from post-mortem studies are easily interpretable and could improve the outcome of pneumonia cases at our hospital. At the same time, they are often difficult because they require gaining consent from a grieving parent. But because I now understood the value of post-mortem studies in protecting all our children from fatal pneumonia, I went home with a stirring in my heart to test the waters of community acceptance. I was reminded of a common saying, “better the devil you know than the devil you do not know,” for despite the challenges associated with post-mortem studies, knowledge of the bugs causing pneumonia today would give us the best chance at saving lives in the future, rather than continuing to treat what we don’t know or what we only think we know.

I am grateful for the opportunity to visit another PERCH research site in South Africa and learn from my colleagues across the continent. I come away from the experience having realized that the value of postmortem analysis in pneumonia etiology studies cannot be understated.  Working in this multi-site project has brought back the nostalgia of my internship year, including the steep learning curve, but also the accompanying fulfillment at the end of each day.

I also have a deeper appreciation that the PERCH study is indeed a landmark study. Working on this project has been an exciting experience and has given me the unique opportunity to be part of the global effort to protect millions of children around the world from pneumonia, and it has shaped my career trajectory towards pneumonia prevention strategies.


Juliet Otieno, MD, is a Clinical Research Training Fellow at Kenya Medical Research Institute (KEMRI) Wellcome Trust Research Programme and a member of the PERCH study team.