Every Wednesday morning, physicians and medical residents in my department gather in the seminar room to discuss morbidity and mortality data of the previous week. Our residents tell us how many patients were admitted to our wards, how long did they stay, how many went home and how many couldn’t. A resident tells a story of one of the patients who died- the case is chosen to identify gaps in resident’s knowledge.

Last week, as we eyeballed the list of deaths, a sense of unease began to brew in the seminar room. Within a week, six patients had died of pesticide poisoning in our wards. Not that we do not see pesticide poisoning often. Every year, we care for about 250 pesticide-poisoned patients – a fifth of them die. But six deaths within a week was a number, too large to be ignored. Why so many people in the catchment area of our hospital poison themselves and why do they die even though they reach the hospital?

People kill themselves for a variety of reasons. For example,a widely quoted Lancet study from China identifies eight risk factors for suicides- depression, previous suicide attempt, acute stress, low quality of life, high chronic stress, severe interpersonal conflict, a blood relative with previous suicidal behaviour, and a friend or associate with previous suicidal behaviour. Indian public health activists and journalists argue that most Indian farmers try to take their lives because they are deep in debt- their cultivation costs are higher, they take loans at very high interests and are not able to pay them off, and they do not get enough returns on the money they spend on their fields. P. Sainath has written a lot on the plight of millions of subsistence farmers in India that led to what he calls the largest wave of suicides in history. I will not go on social and economic issues that forces farmers to commit suicides. Rather, let me figure out why pesticide-poisoned patients run a high risk of death even when they are admitted to the intensive care unit of a tertiary teaching hospital.

Farmers in our area use pesticides in agriculture. These highly toxic compounds interrupt the transmission of messages from the brain to the body- in pests and in people. An important messenger is acetylcholine. After acetylcholine has served its purpose, it needs to be quickly broken down, a task done by an enzyme called acetylcholineesterase. The pesticides inhibit this enzyme. The acetylcholine accumulates in the body and coxes the glands to pour secretions. Fluids fill the mouth, airways, and the air sacs. Patient, frothing from their mouth, are unable to breathe because the poison also paralyses their breathing muscles. Occasionally the victims throw off incessant seizures. Many patients die before they reach a hospital. And those who manage to do require a high quality care in an intensive care unit equipped with infusion pumps, monitors, pulse oxymeters, defibrillators, and ventilators. The treatments are costly. For example, patients have to pay a chemist Rs. 7500 for a 24-hour infusion of Pralidoxime-a life- saving drug. Although our hospital offers the same drug at Rs 2000, most patients do not have enough money to afford even the subsidized treatment.

In our hospital, pesticide self-poisoned patients run a one in five risk of dying. One of five- the number is depressing. What should we do to reduce these death rates? Better drugs, better technology, or better monitoring? Could we be losing these patients because we do not know how pesticides act– how they hit the brain, muscles and glands or how they block the enzymes? The doctors and nurses must know how to protect airways and use ventilators, must have a simple check list of do’s and don’ts; must infuse right drugs in right quantities for a right period; must keep an eye on their patients’ lungs and hearts, must apply high-quality evidence for the care of their patients -consistently, and should work with hospital managers to ensure that patients have access to expensive lifesaving drugs and technology. Could these patients be dying because our physicians, residents and nurses lack experience to handle these emergencies?

Or is it that the knowledge is there but we are not applying it consistently? This reason looks more plausible. Over a two-year period, pesticide poisoning figured four times in our journal clubs. We learned that the one size fits all is a wrong approach- the presentation, clinical course, therapy and prognosis of pesticides depends upon the pesticide the victim chose for self-poisoning. Eddleston’s Lancet article taught us that if patients with pesticide poisoning are to survive, physicians must design specific therapies for them. We had discussed how a simple treatment protocol designed for junior doctors can help us apply evidence- correctly and consistently. We had critically appraised controlled trials of low dose vs. high dose oximes in pesticide poisoned patients from India and Sri Lanka. We were amazed that simple scales could be used to predict outcome – fairly accurately. Although the knowledge existed, and we seemed to be convinced in the seminar room that these journal articles could influence the way we care for our patients, we failed to apply that knowledge where it matters most- at the patients’ bedside.

Sumedh, my resident, is reviewing the hospital charts of all patients poisoned with pesticides over a three-year period. He found that despite evidence to the contrary, our nurses were spending considerable time washing the stomachs of all pesticide-poisoned patients, even though almost all them were past the one-hour threshold. All pesticide-poisoned patients require two drugs- atropine and pralidoxime. Administering atropine is like driving a car in a city at the peak hour- too slow and patients run a risk of getting choked to death. And too fast, patients become agitated and confused, and are “hot as a hare, blind as a bat, dry as a bone, red as a beet, and mad as a hatter”. A review of charts showed that the decisions for starting, stopping or modifying the dose of atropine were neither clear nor consistent. Also, because patients’ relatives could not afford the costs of pralidoxime therapy, physicians tended to underuse the drug. He detected that each unit in the department had its own way of treating poisoned patients- and the doses of atropine and pralidoxime varied widely between as well as within the units. He noted that many patients did not receive diazepam, but phenytoin for stopping their seizures. Finally, he found that hospital charts generated information in bits and pieces- some interlocking pieces of the jigsaw puzzle did not exist in the charts.

Sumedh also narrated his recent experiences -how some pesticide poisoned patients died from what he thinks was an intermediate syndrome – a delayed respiratory failure. Typically patients spend a 3 to 5 days in an ICU where they are successfully resuscitated, intubated and ventilated. Because the doctors think that these patients no longer need an intensive monitoring, they are shifted to a step down unit. Conscious, alert and responsive, with clear lungs, normal heart rates and blood pressures, they seem to be doing well. In fact, they are looking forward to going home. And then they suddenly worsen- their limb muscles, neck flexors and respiratory muscles turn weak. They are not able to breathe properly. Their oxygen saturations go down. Secretions clog their airways and lungs. Some patients die before their worsening is detected and some die in the ICU- even though they are supported on ventilators and are re-infused with antidotes.

Pesticide poisoned patients require just two life-saving drugs (atropine and pralidoxime); only two organs to be carefully observed (heart and lungs); just two equipment to see them through the stormy hospital course ( infusion pump and a ventilator) and only a simple scale (Glasgow coma scale) to judge how would they fare. And yet, we were failing them. Was it ignorance or ineptitude- Atul Gawande should tell us.