The typhoid case was one of several that slipped through the cracks due to system failures in that first week.

In that same stretch, we also had an 18 year old woman admitted with abdominal pain and fevers, with ultrasound showing a likely liver abscess (walled off infection pocket). She was stable, but with a high heart rate and looked fairly sick. We made a note to watch her closely and get her a CT scan as quickly as possible so that surgeons could get involved if the pocket was big enough to drain. The next day we heard that she was found in her bed deceased by nursing staff overnight. When Adeline and I heard about it, we were shook, and went straight to the chart to see what had happened. Opening the file, we saw she was reviewed after a clinical status change some time in the evening, and by 4:00AM she had died. Documentation of the events overnight left a lot to be desired. The attendings were not made aware until we found out she died the next day.

We had another case presented at morning report about a 28 year old male who came to the ED around 10:00AM with concerns for infection and documented hypotension on arrival, essentially in septic shock. The medical student clerked the patient in the ward that night around 10:00PM and the patient was still in septic shock. The patient then died overnight. At no point was any attending made aware.

These three cases highlighted a clear need to improve the frameworks by which these patients were monitored and managed. Despite being a limited resource setting, we still had enough tools to do better: a CT scanner, the capacity to deliver ICU level care if needed, and the availability of experts that could provide more input or get the patient referred to a higher level of care by ambulance in Kigali. Mainly, it seemed like there was a failure to take action for a clinically worsening patient. I think we fail to appreciate this in the high income world, because we have such profoundly well trained nurses that will sound alarm bells immediately when they notice people getting sicker and a culture that encourages residents to take action and ask for help when out of their depth. Here, due to a variety of factors, that wasn’t happening. Like I have said in this blog before, the issues of poor care in resource limited settings run so much deeper than just resources themselves, we are talking about culture shifts in the practice of medicine: accountability, agency, and belief. Experience has taught me that this can never come from someone like me with a Western accent giving courses in point of care ultrasound. It has to be shown in the trenches, side by side with local staff demonstrating that we can uphold a different standard.

So for the rest of my time with this cohort, Adeline, Zahir, Alex, and I made it a point to hammer this into the clerks: When patients are sick, we have to take action. After we take action, we re-evaluate promptly and take action again, with the cycle repeating until the patient has been sent to the highest place they can reasonably go, or dies with us at the bedside doing our best.

As fortune would have it, just a few days later, we got such a case.

Like always, it started at morning report. It was one of our lady students.

‘I would like to present the case of a 38 year old male who presented to the emergency yesterday with a history of one week of shortness of breath with associated cough and fevers. He has a known occupational exposure with a background diagnosis of silicosis (a disease from silica inhalation that causes thickening of the air sacs in the lung). He presented yesterday due to shortness of breath that was progressive and worsened with activity as well as an associated cough and wheeze. He denies active chest pain or bloody sputum. The patient also reports fevers started on the days leading up to admission with no vomiting, diarrhea, headaches, or new rashes. On examination, the patient was ill-appearing and in distress, with a heart rate of 115 and breathing at 30 breaths per minute on 15 litres of oxygen saturating around 92%. His blood pressure was in the normal range and the patient was without fever.’

‘You said how much oxygen?’ I asked.

‘15 litres,’ she said.

‘And he was breathing at 30 breaths per minute?’ I followed up. ‘Saturating 92%?’

‘Yes.’

Several of the attendings in the room and I exchanged looks.

This guy was sick as shit and I could nearly guarantee he was not being appropriately managed.

‘On exam, the cardiac exam was notable for present heart sounds and no murmurs. His lungs were found to have coarse crackles bilaterally with wheeze and partially absent breath sounds at the lung bases. His abdomen was soft without tenderness. His peripheries were warm with pitting edema. The patient was started on 1L maintenance fluid, with labs sent but still pending and ordered for antibiotics with a preliminary diagnosis of hypoxic respiratory failure secondary to pneumonia and silicosis.’

Again the attendings all exchanged looks. Then the poor student bore the brunt of our grilling as we dissected the diagnostic evaluation and plan. In reality, the students just saw the patients at night and followed what was written and ordered by the medical officers of duty, so none of these were criticism of the student’s clinical thinking, although we always made a point to encourage them to be inquisitive of the plan so that they understand the rationale behind all of the decisions.

The truth is that by now we had learned that these patients were likely to be mishandled in the emergency room where there was only one medical officer and no attending supervision. The reason for this is that the division of internal medicine was under the University umbrella but the emergency room was not, so the medical officers there had no accountability to us. In the medicine department, we also weren’t technically staff attendings, but instead internal medicine ‘consultants’ who bore no legal responsibility for the patients. We had a good relationship with the medicine division and the officers listened to us, but ultimately they didn’t have to. This is part of the nuanced dynamics of having university staff in a hospital that doesn’t belong to them.

Among us at morning report we also had Dr. Pascal, the internal medicine staff doctor, not employed through the university. He usually took on a more managerial role but was in theory the attending responsible for all of the patients admitted to medicine. Whenever we needed to escalate or make things happen, that all moved through him. He was Rwandan, probably in his late forties, with a smooth head and big smile. He spoke with a sharp but intelligible accent and made several teaching points on the case before we concluded morning report.

‘We should go see this patient now,’ he said to the clerk. ‘He is very sick.’

So the other students scattered to do their pre-rounds and I tagged along with Zahir, Alex, Adeline, and Pascal to go see our patient still in the ED.

Boy did he look rough. Sitting up in bed, from several metres away we could see his chest heaving, with pronounced collarbones that tethered strained neck muscles to a protruding chest. He was sweaty, with mouth open in a face mask that hissed air at 15 litres per minute.

The clerk assigned to him was name Anaise, and we examined the patient together up close. His lungs were definitely crackly on both sides, but he also had a fairly pronounced wheeze that made it hard to hear his heart. His neck veins bulged prominently and his legs were swollen to touch, leaving a big pit (pitting edema, as we call it, a sign of excess fluid).

Zahir, Pascal and the others were nearby as we surveyed the patient together, he looked like he was headed towards needing a breathing tube.

‘Reasonable enough to get a blood gas and a chest X-ray?’ I said.

‘Oh heck yeah,’ Zahir answered.

In the meantime, I pulled Anaise aside and we started breaking down the case.

‘When patients are this sick I like to take an organ-based approach,’ I told her. ‘So let’s start.’

One by one, we talked about all the systems from head to toe and I let her craft a plan with some direction. For the moment his neurologic status was intact. His cardiac function seemed potentially compromised with the distended neck veins and fluid overload, so we made a plan to collect an electrocardiogram and give the patient diuretics to help remove the excess fluid. His lungs were the big issue, so we ordered inhalers and steroids for his wheeze, and kept him on the right antibiotics for a potential pneumonia. His GI system didn’t seem like a big issue, but we ordered liver enzymes to see if he had any organ dysfunction and watched for good poops to prevent any obstruction. The kidneys were another hot topic, his labs taken the day before revealed a kidney injury, which could have been related to sepsis or excess fluid, so we agreed to place a catheter and watch his urine output. With regards to his blood, a count from the day before showed normal white cells, and relatively normal findings otherwise. From the infection standpoint we would have to check some more studies for a source of sepsis if that was driving his process.

‘Most importantly,’ I said. ‘We don’t just write these things down and walk away. I would like you to get the nurse, have him or her stand with you and outline the whole plan with a priority on starting the diuretics and collecting the studies followed by the catheter. Then we will come back in 30 minutes to re-evaluate the patient. This guy is sick. He is an ICU level case right now, which means the other less sick cases can wait. Does that make sense?’

‘Yes doc,’ she said.

With an empowered stride, she did just that, and even pulled in the emergency room medical officer.

In short order the portable X-ray machine was wheeled in and snapped an image. Watching it come up on the screen, Zahir and I both felt the same way.

‘Certainly looks like patchy infiltrates,’ he said.

‘Multi-focal definitely, hard to say from what, though,’ I said. ‘Prior silicosis with bilateral pneumonia or congestive heart failure? Or both even?’

‘Can’t really know at this stage,’ he said.

Shortly after we got the results of the blood gas and the results were alarming. His carbon dioxide levels were above normal.

When we hyperventilate, that increases lung ventilation and clears carbon dioxide from the blood stream. Because carbon dioxide is acidic, clearing it at a higher rate leads the blood to become alkalotic (the opposite of acidic). In patients who have low oxygen and hyperventilate to compensate for this, the carbon dioxide should be low. However, if patients breathe so heavy for so long that their respiratory muscles start to tire, the carbon dioxide levels start going up. This can be a sign that the patient is failing to meet the demand to keep their lungs oxygenated. Once patients tip over the edge, the carbon dioxide levels skyrocket and oxygen plummets.

‘I think we need to intubate,’ Zahir said.

Despite not having a formal ICU, Butaro Hospital does in theory have an ICU program. The ICU was currently closed because the space usually used by the ED was under renovations and the ED was using the ICU space instead. So we did have the capacity to intubate and run ICU level cases, it was just technicalities of bedspace that prevented us from having a formal ICU team with dedicated ICU nurse staffing.

‘Pascal,’ I said. ‘Have we got that anesthesia doctor here?’

‘Yes, he should be. We will call him and they can come and review,’ he answered.

‘Sweet!’

When the anesthesia doctor arrived we discussed the case with him and he agreed to co-manage the patient as an ICU case. This gave us some liberty to go to the medicine ward and round on the other patients.

‘You see?’ I told Anaise. ‘We identified a sick case and rang all the bells. Now everyone in that ward knows how sick he is. Nursing knows to watch him close. Anesthesia will review him for intubation if needed. If he needs to be transferred he will be transferred. If we just wrote things in the chart and walked away, he sits there unattended for hours and hours like some of those other cases and nobody takes notice until he’s bordering on a cardiac arrest or already dead.’

‘Yes, I agree,’ she said. ‘I hope he will do okay. I think we are giving him good care.’

When we circled back a few hours later, the patient actually looked better. The diuretics were taking effect and his breathing was more settled, which spared him getting a breathing tube. Since this case was on a Friday, I went back to Kigali to do some work at another hospital and didn’t see him again until the next Thursday. By then he had been diagnosed with new heart failure and was getting fine-tuned before discharge

That’s what it’s all about, aint it?