Admitting patients to the hospital is my favourite part of internal medicine. The first 24 hours of a patient’s journey in the hospital are typically the most active, as medical teams deliberate the diagnosis and management approach before committing to a plan. For the internal medicine physician, this is the time period that allows you to act like a detective and put your training to use.

Hospitals have different ways they distribute admissions, some have every team admitting all the time, while others designate specific teams to specific days. At Princess Marina, the teams admitted in a cycle, such that one team took all the incoming admissions for 24 hours at a time. So, just after rounds on Tuesday of my second week, Neo and I walked to the emergency room to gander at the board and see if anything was coming our way.

There is always an air of madness in the Emergency Room, which is part of why I like it. Marina’s was arranged in a large square, with the main workspace in the middle and patient beds along the walls. When the beds were full, patients sat in the hall on chairs or lay on the ground. The workspace was littered with CDs of scans, X-ray films, charts, binders, and paperwork of every colour. Walking into an air-conditioned jolt from the dense heat, we walked up to a whiteboard with numbered columns and rows.

‘We just look for the ones that say medicine review, and then we evaluate for admission,’ Neo told me.

I saw we had four waiting. Easy peasy.

The first two of the cases were straightforward, a heart failure exacerbation that would be fine and discharged later that day, and a patient with a kidney injury from a urine blockage that needed urology follow up after getting a catheter placed. The other two, not so much, and they were both sick as shit.

One of them was a gentleman in his early seventies. He had been brought in by his family for strange behaviour at home and worsening confusion. On arrival, he had low oxygen on a pulse oximeter. The Emergency team had not been able to discern much information due to his poor mental status, although they did know that he was a patient with HIV with an on-and -off history of taking HIV medications. They got some initial lab tests and a chest X-ray, which was what prompted them to call us.

On my evaluation he was lying in bed, knees bent up. His eyes were open but empty, and he was squirming with no specific intent. His ribs were protruding, so much so that I couldn’t press my stethoscope flat against his skin to hear anything in his chest. I felt a strong pulse, with no notable heaving in his chest. When I tried to sit him up to listen to his lungs, he was unable to cooperate, and turning him over was an ordeal. He wasn’t having a particularly bad cough, nor did his breathing look laboured, but his oxygen was truly low; in the mid-80s when measured. His abdomen was taught all around from firm abdominal muscles and he had no large organs or pain. When I tried to lift his head so that his chin would touch his chest, his entire torso came up with it, as he kept his neck straight no matter how hard I tried to coax him. It was difficult to tell whether his neck was stiff, or if he was voluntarily resisting my movements. Overall, he didn’t look acutely septic, but his poor mental status in combination with confusion and neck stiffness was concerning. Looking at the monitor, his vital signs were relatively normal, which was good.

I found his chest X-ray and placed the film on a light box, revealing a pretty honking apical pneumonia on his right side, extending into the middle portion of his lung. This distribution was typical of the leading infectious killer worldwide, tuberculosis.

Tuberculosis, or TB, is a bacterial infection that dates back to prehistory. Archeologists have isolated TB in the bones of Bison dating back to more 15 000 years before Christ, and in the spines of mummies from ancient Egypt. The earliest writings of TB-like disease in humans date back to Indian and Chinese antiquity. Its prominence rose and peaked in the 1800s, where it caused 25% of global deaths and generational epidemics in Europe, where it was referred to as Consumption, becoming more widespread across the urbanised poor. This led to widespread medical research, culminating in the isolation of the TB bacteria by the German physician scientist Robert Koch. The bacteria gets its name from the latin word tuberculum meaning ‘lump,’ in references to the swollen lymph nodes typical of patients with the disease. Once discerned to be an infectious disease, common treatment for close to a hundred years was admission to sanatoriums, healthcare facilities isolated from the uninfected with an emphasis on warmth, good nutrition, and fresh air. Despite this, mortality at sanatoriums still hovered near 50%. Despite having been eradicated in most of the developed world after World War II with the introduction of antibiotics and improvements in public health, it remains the second leading cause of infectious death worldwide in 2022 (second to Covid-19 and above HIV/AIDs). Per the WHO, it is thought that one quarter of the world’s population lives with TB either in the active or dormant form of the disease today. I have seen only a handful of cases of tuberculosis in the US, all at the Penn refugee clinic. Many physicians in the US will never see a case of it in their career. Here, I learned fairly quickly that TB was still alive and well, and it can present in a variety of bizarre ways.

I had to presume this gentleman was not actively on his HIV medications, which left him exposed to infections. Even though apical pneumonia is most commonly TB, it’s important to consider that atypical patterns of regular pneumonia are also more common in patients with uncontrolled HIV progressing to AIDS. He could have had mental status changes from sepsis or pneumonia alone, so we had to keep an open mind. Either way, he was sick and I was ready to throw everything at him.

I discussed the case with Neo and we agreed. Antibiotic coverage for regular pneumonia as well as TB treatment. I talk about this case as if I called the shots, but in reality I leaned heavily on Neo for this one. She had seen and treated more HIV and TB than I have read about, and so I got her to walk me through what agents to give and how far to expand our coverage.

‘I would treat for pneumonia, sepsis, and TB,’ she said. ‘We can start prophylactic fungal coverage while we wait for HIV tests, but this CXR pattern is not typical for fungal pneumonia despite it being more common with HIV. For the possible meningitis, we don’t see much viral meningitis here, and even if we did, we don’t have IV agents for Herpes or Cytomegalovirus available and would have to wait up to three months. If we really suspect TB meningitis, we can see how he does with this while we wait for lumbar puncture before we commit to steroids given we’re still not sure.’

‘Sounds like a plan,’ I said.

In the end we admitted him, and a member of our team was able to get more personal details from the family. He had been acting strangely and losing weight for a few weeks after stopping his HIV treatments. It was not uncommon for patients here to seek herbal medicine alternatives to treat their diseases, and he was also a herbal doctor. Even once he got admitted, it proved increasingly difficult to treat him as he would spit out the medication in his confusion. It was only after a few days that we devised a way for his family to come in and feed him lunch with his medicine. His spinal results came back inconclusive, but failure to improve prompted the team to start him on steroids and extend his treatment course for TB meningitis. He got better, continued on his TB therapy and was again started on HIV treatment some time later.

The other patient, sadly, did not fare so well.

He was younger, forty-two. He worked in a mine, although he had stopped working around 5 months ago when he was diagnosed with HIV and TB together. At that time, he was started on treatment and improved, but then for reasons unclear he had a lapse in treatment for about three months. He was called out to us as soon as he walked in the door for sepsis and a relapse of TB lung disease.

Looking at him in the bed, he was thin, but not pure bones like my other guy. Additionally his abdomen was swollen. He was on a breathing mask, with respirations looking laboured. His heart rate was elevated but blood pressure was holding. He was sweating profusely, and had panic in his eyes. He spoke decent English, so I was able to ask him more about his symptoms. He told me they started about a month ago, first with watery diarrhea a few times a day. Then came the shortness of breath and feeling terrible, which was accompanied by night sweats and fevers that had persisted. On the day of admission, he started having pain with breathing and wound up here at Marina. When I examined his heart, I found not much of consequence. I tried sitting up to hear his lungs but he couldn’t maintain a seated position and slouched back down immediately after rising due to weakness. When I listened as best I could, I heard turbulence from his deep breaths but no focal crackles or empty spaces. In his abdomen, he had a big liver that was sore to touch. He looked awful, and the medical student that accompanied me looked like he had seen a ghost.

 ‘Frame this in your mind,’ I said. ‘This is sepsis. We have one job right now, and that is fluid and antibiotics, everything else, put it aside.’

He nodded, jotting notes down in his book.

We got a hold of his chest X-ray, and it was bad. He had the apical consolidation on one side, and what looked like a cavitary lesion on the other. Near to the lesion he also had a pneumothorax, which is when air enters the space between the lung and the chest wall. This can be highly dangerous, because if the pneumothorax expands, it can compress the lung and heart, which can be fatal. His pneumothorax was small, not big enough to warrant inserting a tube to drain the air, but it may have been contributing to his breathing and pain. Still, despite how bad he looked, he was young, and sepsis could be treated. I had hopes that he would do okay.

Again, Neo and I went over the plan, involving the medical student in our discussion. His name was Ron, he wore horn rimmed glasses and spoke articulately. He was very thorough, like so many of the trainees here. As was the standard process, he would take charge of the admission under our supervision and follow the patient for the duration of their stay. I remember he tried to present the case and kept getting tangled up in all the problems.

‘This is a complex one,’ I told him. ‘When there are so many problems, I find it easier to break up in terms of system. Try that and see how it feels, start with neuro.’

‘Ah, okay,’ he said. Again, he jotted down in his book and took his time before starting again, and did much better.

‘Good!’ I said. It’s nice to see the wee ones succeed.

We put together a plan for his many problems. He had the clear pneumonia, with the cavitation and pneumothorax, this would require a sample of his sputum and close monitoring for any sign of decompensating. If he got worse suddenly, we would have to consider a chest tube or needle to decompress. His diarrhea could very well be related to an infection, parasitic, bacterial, or TB, so a sample of his stool would go to the lab. He would need fluids, antibiotics to cover all sources, including TB and bloodwork to follow up first thing in the morning. As of right now he was stable, but ripe to get worse.

‘Remember Ron,’ I said to the student. ‘This guidance has to be very clear. If he suddenly clinically deteriorates, we have to be suspicious of the pneumothorax getting bigger. That needs to be written at the front of your note. If somebody gets called to see him during the night, we have to keep this in our minds.’

‘Yes, yes, doctor,’ he said.

The next morning I saw the patient again before rounds. He was in a separated part of the medical ward under isolation conditions and looked about the same, maybe slightly better from the fluids and the antibiotics starting to take effect. We saw him as a team and the attending methodically dissected every organ involved and added more items to our plan in light of the blood work showing both a kidney and liver injury. We would have to get a dedicated CT scan of the chest to see the story with the cavitary mass, a liver ultrasound, and a blood sample under the microscope as he had low cell counts, possibly attributable to sepsis, uncontrolled HIV, or something else. Additionally, he wanted us to get an ultrasound of his heart looking for an effusion—a subtle enlargement of his heart on the X-ray prompted this. It was a detail that I missed—He was the first patient we saw, so all of these things would have to wait until after rounds. That’s just how it is.

Some hours later when rounds finally finished, the attending told me he had his own ultrasound probe and wanted to learn how to use it. It was the same model I had used in Uganda to train the staff there. We broke off from the team and collected the probe before heading towards the isolation room to get a look at his heart and liver. I remember he walked in first and I lagged behind, grabbing an N95 mask from the wall outside before coming into the room. Just inside the door, I set the ultrasound probe down and took off my glasses and surgical mask to put on my N95. Just as I was doing it, I heard the door open behind me and slam shut followed by loud footsteps down the hall. I looked up to see the attending was gone, and our patient was lying with eyes and mouth open, still as a stone.

Fuck, I thought to myself.

I dropped everything and went straight for his neck. Pressing firmly I felt nothing. Taking three steps out towards the door, I poked my head out and heard the attending shouting.

‘Resus! Resus!’

Back I turned towards the patient and dropped the bed to start CPR. With my first compression I felt ribs break, and I focused on my speed. One hundred beats per minute, every time.

About ten seconds in I realised I didn’t have a mask on, and looked right to see my glasses, the ultrasound and both masks laying on an empty bed by the door. As far as high risk exposures to TB go, doing active CPR on a patient with pulmonary TB sepsis is about as high as you can go. In that moment though, I wasn’t thinking about me, and stayed on the chest. Around 30 seconds later more people started coming into the room, the attending with the other residents and some medical officers. Nurses also filtered in slowly, all of them diligently putting on N95s by the door before stepping in. The attending took the bag mask and started bagging. By now I had done almost a full cycle of compressions getting close to 120.

‘I need you to take over after pulse check,’ I said to the medical officer in front of me.

He nodded.

I pulled back and one of the nurses handed me an N95. I then slid to the foot of the bed. I run a lot of codes in Philadelphia, and hold myself to a high standard to run them as best I can. It wasn’t clear to me if anyone had taken ownership of running this one.

‘Who is keeping time?’ I asked. To be honest I kind of shouted.

The nurses seemed to look at me rather oddly. As if to say ‘why is this guy shouting’. Nobody really said anything, so I presumed that to mean no.

‘I am keeping time,’ I said aloud. ‘Can we give the first dose of epinephrine please?’

I stepped away from the foot of the bed and to the side, I now designated myself as the time keeper, and didn’t feel particularly liked as the code leader. I didn’t want to step on any toes given my foreigner status. Watch in hand, and notebook open, I documented and called out the intervals.

‘Pulse check in 30 seconds.’

‘Next dose of epinephrine is due.’

‘Pulse check in 30 seconds.’

Some of the residents were getting the defibrillator off the code cart. It looked completely unlike any I had seen before, tall, with a built-in monitor and two large metal hand pieces. It was an older style that required the operator to manually press the pads onto the patient’s chest to deliver the shock like in the TV series I remember as a kid. Dr Kenny, the same chief that had showed me around when I started, rubbed the pads together with gel and pressed them to the chest. A line came across the monitor.

‘Looks like asystole,’ one of them said.

There are many forms of cardiac arrest. Not every arrest is when the heart stops altogether. Depending on the electrical activity of the heart during an arrest, you can potentially correct the abnormality with a shock or special type of medicine. For this reason, getting the patient on the monitor is important to determine if it’s a rhythm that can be shocked or not. In this case, a shock would do nothing so we held off. As the code continued I looked up and saw a nurse with a notepad.

‘You are keeping time?’ I asked.

‘Yes,’ she answered.

Okay then. I was no longer the time keeper. I tagged in for another cycle of compressions before stepping back. We were now probably ten minutes into the code. I went through all of the possible causes, we had gotten a blood glucose already which was normal, he wasn’t bleeding that we could tell, and his electrolytes were all relatively normal when last checked. I turned to the attending.

‘This patient has a pneumothorax,’ I said. ‘He was clinically stable this morning before this sudden arrest, I think it would be reasonable to try a decompression. Have you any thoughts?’

His head tilted. I don’t know that he liked the idea.

‘Can try,’ he said.

I turned to the medical officer next to me. He was assigned to my team and we got on well.

‘Have we got material for a chest tube anywhere nearby or do I sound insane?’ I asked.

‘No no… We don’t have that,’ he said.

I thought to myself.

‘Okay, needle decompression it is.’

I opened the code cart and found the largest cannula I could. I held it in my hands before looking around. I had never done a needle decompression before. In theory, it was no different than placing a chest tube or tapping an effusion, procedures that I have done, but still, nobody likes to experiment on a human being. I turned to the medical officer next to me.

‘Do you usually go for the 5^th and 6^th rib in the axilla or the 2^nd and 3^rd anteriorly?’ I asked. This was in reference to the two most common targets for decompression.

He had no idea.

‘I don’t know Juan, I am sorry.’

With all the commotion happening on the patient’s chest I opted to go for between the ribs at the side and squatted next to the bed. His torso was shifting up and down the CPR, so I had to wait. I looked at the attending and confirmed with him again that we felt this was the right move. He nodded.

‘Pulse check,’ I heard.

Feeling above the sixth rib, I inserted the needle perpendicular into his chest. I went in without resistance to a fair depth; I felt like I was certainly in the right place.

Nothing happened. I pulled the needle out of the cannula, and then removed the cannula. No pneumothorax to decompress. I could do nothing more

CPR had restarted now, and that was that. I stood back, and we called the code off after twenty minutes. The room slowly cleared.

I thanked everyone for their assistance and stepped out. The attending hung around for a few minutes before leaving, he would call the family. I had to fill out some paperwork for the arrest protocol. This wasn’t the first or the last patient I had die under my care, or in my hands for that matter. I tried to do what I usually do, pay silent respect and take a moment for reflection. Truth is, the day goes on, and we had a full list of other patients that still needed care. I then found the rest of the team.        

‘Right then,’ I asked the students. ‘What do you guys want to learn about today?’