Publications
Partners
- Autopsies shed light on what people actually die of.
- But it can also play an important role in helping us to better understand disease.
- Spotlight delves into some of the fascinating complexities of tuberculosis-related autopsy research.
Figuring out how many people in South Africa die every year of tuberculosis (TB) is not straight-forward.
On the one hand, Stats SA's frequent mortality reports put the number at under 30 000; on the other hand, the World Health Organisation (WHO) estimates that it is over 50 000.
While this may at first glance seem like a large discrepancy, there is a simple explanation. The Stats SA figures are based on what is written on death notifications, and these notifications very often do not tell the full story of what a person died of. The WHO estimate is derived using mathematical modelling that triangulates estimates based on several data sources.
Looking at the numbers from studies that determine the cause of death (or what people actually died of) is one of the ways we know that relying on death notifications result in an undercount of TB deaths. Such autopsy studies have consistently found that many people had undiagnosed TB at the time of death and that the undiagnosed TB was often the actual cause of death.
One review study published in the journal AIDS concluded that "in resource-limited settings, TB accounts for approximately 40% of facility-based HIV/AIDS-related adult deaths" and that "almost half of this disease remains undiagnosed at the time of death".
According to WHO figures, of the estimated 280 000 people who fell ill with TB in South Africa in 2022, over 65 000 were not diagnosed.
Importance of autopsy research
Dr Muhammad Osman, Academic Portfolio Lead and Senior Lecturer: Public Health at the University of Greenwich, tells Spotlight that it is important to do TB autopsy studies because it enables us to identify TB that was not diagnosed during life - and this helps us understand the true burden of the disease.
READ | Huge TB vaccine trial starts in SA. It could deliver first new TB jab in over a century
Osman says identifying TB at autopsies has significant benefits. He says by overlaying health seeking behaviour (how people visit clinics), we can identify missed opportunities for TB screening and design interventions to improve screening for TB.
"We could trace family contacts of the deceased and offer TB screening and prevention. This is not taking place at present," he says.
Osman and his colleagues published a paper in the International Journal of Infectious diseases in 2021 looking at TB in people with sudden unexpected death (SUD) in Cape Town. They found that active TB was identified at post-mortems in 6.2% of the 770 cases they studied. More strikingly, in around 92% of those cases the TB had not been diagnosed while the person was alive.
Osman says that these days there is an increasing awareness of undiagnosed and untreated TB. He points out that new interventions to improve TB testing and diagnosis have been implemented such as targeted universal testing — an approach by which people who do not have any TB symptoms, but who are considered to be at high risk of TB, are routinely offered TB tests.
He says these days healthcare worker risk is considered more carefully and he stresses the importance of protecting forensic and pathology teams. (Forensics focuses on determining the cause and manner of death, while pathology is the study and diagnosis of disease through examination of tissue, cells, autopsies, and so on.)
Closing the gaps
Osman says their study also identified a gap between the pathology services and access to routine health service records.
"We thought that this is an essential gap to close - the forensic/pathology services need access to routine health service. For a limited number of these deaths we were able to match their records to the public health clinic and hospital records - and many of them had contact with the health services in the six months before death," he says.
"If forensic pathologists are given full access to the health records, they would know the timing of previous TB and the treatment outcomes of those episodes. The lung changes seen with TB are different in the case of active TB and healed/recovered TB. There are well documented macroscopic (what's is seen by the examination) and microscopic (seen through histology and microbiology) findings," says Osman.
A complex disease
The study of TB is complicated by the fact that TB can occur at several stages on a continuum and can impact several different parts of the body.
Professor Threnesan Naidoo, research pathologist at the African Health Research Institute (AHRI), tells Spotlight that when people think of TB, they usually think of the person who's been coughing for a few months, loss of weight, loss of appetite, having night sweats, and maybe coughing up some blood.
"But there's a journey to that point and then generally beyond that point, and clinically, there's a continuum of the disease. We refer to it as latent disease, subclinical, active and then healed TB," he says.
It is an area in which things are changing fast – a paper published in the Lancet medical journal last week proposed dividing TB into five stages.
ALSO READ | New TB drug shows promise, but experimental vaccine disappoints
Naidoo says autopsies provide an opportunity to study TB at different stages (latent, subclinical, active, healed) especially when someone with TB dies of another cause. He says they can encounter people at any stage along the TB continuum because at any point someone could be shot, stabbed, or involved in a motor vehicle accident.
"You (pathologist) have a unique opportunity to study the effect of TB on cells and tissue physically under a microscope and not through imaging (x-ray)," Naidoo says.
Autopsies also presents the opportunity to look at TB disease not only in the lung, but also the brain, thyroid gland, kidney or urinary system since TB has the capacity to spread everywhere, explains Naidoo.
"Autopsy gives you the opportunity to study TB everywhere," he says. "Clinically (when someone is alive), you don't go about investigating the entire body - neither is it practical nor feasible or safe. But [with an] autopsy, you're examining the entire body anyway. We study TB in totality," he says.
How it is done
The standard manner of doing an autopsy involves a thorough examination of the body.
Naidoo explains that the process starts with an external examination to document injuries, marks and other physical characteristics that are visible. The internal examination involves dissecting organs, tissues and body cavities to identify any abnormalities or signs of disease. Samples may be taken for further analysis, such as toxicology tests, histological examination, or TB research.
Any findings from the samples, Naidoo notes, must be interpreted taking into account changes that occur in a dead body.
"[In] the living, you know, it's a living person and they're able to do things and you're able to see things on imaging (X-Ray), but in the dead you have to account for the fact that the person has now demised and certain changes occur after death."
Autopsy study at UCT
An ongoing study at the University of Cape Town is exploring the role of lymph nodes in the spread or containment of TB disease by looking at tissue of the deceased.
Much TB research so far has been done on animals and not on humans, points out Dr Virginie Rozot, research officer at the South African TB Vaccine Initiative (SATVI) and co-principal investigator of the UCT study.
"We have great non-human primate and great mice studies that try to underline the mechanism of the disease progression. However, animal models are not a true reflection of what happens in humans.
"With humans, it's really complicated to point to where the disease might happen and to understand the immunology happening in the tissues where generally the disease happens," she says.
"For the longest time in these human studies, most studies have been done in the blood and what is happening in the blood has been taken to correlate with what is happening in the lung."
READ MORE | To end TB, we need to provide vulnerable people with food
In short, autopsies allow researchers to look directly at lung, brain and other tissue in a way that simply isn't feasible in living people.
"So, the only way you can actually access tissues is to do post mortem studies. Post mortem studies have been happening since the beginning of last century. And they were like fantastic studies, but the tools were not the same as we have today. I think that should come back to the front of the scene of research because then you can ask all the questions we've been trying to answer on what is happening in the tissue by looking into the blood," she says.
"Autopsy allows us to study the exact part we want to study, not just the blood."
Collecting samples
In collaboration with the Western Cape Forensic Pathology Service, UCT has created a postmortem sample collection platform to help with TB research.
By leveraging the Inquest Act of 1959, which states that people that die of unnatural causes must undergo a medico-legal investigation to determine the cause of death, Rozot and her team come in to conduct a post-mortem to get their samples. They aim to do the post-mortem in less than 24 hours after death.
Since starting this study about eight months ago, they have done 125 autopsies, with a consent rate of 64%.
"I think our consent rate is incredible. We are still putting together our findings to determine how many cases of TB we have found so far by looking at autopsies," says Rozot.
Representative samples
Dr Laura Taylor, forensic pathologist at the Western Cape Forensic Pathology Services,, says the bodies that they look at, in line with the Inquest Act relating to unnatural deaths, are representative of people in South Africa.
"However, they are not exactly representative of the entire South African population because there are certain socio economic groups that are more likely to die of unnatural deaths due to increased prevalence of trauma and violence in their communities," she says.
ALSO READ | SA has one of the highest TB burdens in the world. Now a researcher is eyeing AI as a new weapon
Because there is no central database, Taylor couldn't say how many cases of TB they find among the deceased.
"[T]here are autopsy records or reports which are written for each case, but there is no central database for TB specifically detected [through] autopsy," she says.
Forensic autopsy and other diseases
Rozot and Naidoo share the view that, if done well, TB autopsy studies can help shed light on other diseases.
The value of this information is that people dying with or from TB will also have any of the other conditions, such as hypertension, HIV and diabetes, Naidoo says.
"You can work out all those variables… [people] don't just come with diabetes - the diabetes changes the face of TB, HIV changes the face of TB, and TB changes the face of those diseases as well. So, the complexity of it becomes something that we need to pay attention to, and look at all the common variables, like the association of TB and HIV is a big one.
"So studies might look at HIV infection and how it may affect TB and vice versa - same with diabetes, hypertension any of the other non-communicable diseases as well," he concludes.
*This article was published by Spotlight – health journalism in the public interest. Sign up to the Spotlight newsletter.
