The Government of Colombia's Procedures for Handling Claims of Colombian Citizens That Their Health Was Harmed or Their Licit Agricultural Crops Were Damaged by Aerial Eradication
 

¹Report on two deaths in La Argelia (La Hormiga, Valle del Guamuéz) allegedly caused by glyphosate. 24th Army Brigade Intelligence Division, November 25, 2001.

According to the same report, another child in La Argelia experienced diarrhea, which improved initially then worsened, accompanied by respiratory failure. He died the next day at the same hospital.

The family of the 11-month old deceased child and the municipal representative attributed his illness and death to aerial spaying of coca in La Argelia township, La Hormiga Municipality, where he resided with his family. The family of the 4-year-old did not, but the hospital apparently reported both cases to the Army at the same time.

The author of this report was retained by the U.S. Embassy, Narcotics Affairs Section, to investigate the deaths. He traveled to La Hormiga by plane on Wednesday, December 5, departing at 09:00 hours and arriving at 10:30 hours. There, he met with the municipal representative, Mr. Leandro Antonio Romo-Pantoja, and the Director of the Sacred Heart of Jesus Hospital, Dr. José Victor Pérez (both of whom had been in contact with the families of the deceased) and reviewed the children's medical records.

According to the municipal representative, both cases were preceded by the death of fish and chickens in the villages of El Tigre, El Paraíso and Maraveles, at about the time of aerial spraying on November 13, 2001.

The commercial formulation of glyphosate used in the illicit crop eradication program consists of glyphosate, water, and a surfactant. This formulation is mixed with water and a Colombian-manufactured and approved surfactant, Cosmo-Flux 411F. A surfactant, which is like a soap, helps the herbicide penetrate the leaves of a plant.

The dose used in the aerial eradication program for coca spraying contains 10.4 liters (2.75 gallons) of the commercial glyphosate formulation per hectare. The dose of the active ingredient, glyphosate, is 3.74 kilograms (acid equivalent) per hectare. This is equivalent to 374.4 mg of glyphosate per square meter.²

There is a large body of scientific literature indicating that glyphosate is of low toxicity for humans, animals and the environment.³

3. CASE REVIEW

During the interviews with the officials mentioned above, the complaint filed with the municipal representative by the child’s relatives was reviewed, as was his clinical history. It says, in summary, that a nursing child 11 months old was admitted to the hospital in a state of shock, and that his relatives had reported that his symptoms initiated 24 hours earlier, characterized initially by diarrhea and vomiting on multiple occasions and instability after exposure to spraying for fumigation. According to the municipal representative, these symptoms were preceded by nasal and ocular itching, inability to maintain posture and ocular deviation. However, this information is not included in the hospital’s emergency care record, according to which " the symptoms worsened and his condition at the time of admission to the hospital did not respond to treatment." ⁴

²Parra, L.E., (Consultant to Narcotics Affairs Section, US Embassy, "Overview of Risks of Exposure to Glyphosate," July 2000.
³Environmental Health Criteria No. 159, Glyphosate, International Programme on Chemical Safety, World Health Organization, Genève, 1994; Williams, G.M., Kroes, R., Munro, I.C., "Safety Evaluation and Risk Assessment of the Herbicide Glyphosate," J Regulatory Toxicology and Pharmacology, 31: 117-165, Academic Press, 2000; Giesy, J.P., Dobson, S., Solomon, K., "Ecotoxicological Risk Assessment for Glyphosate Herbicide," Rev. Environ Contam Toxicol, (Sic) 167: 32-120, Springer Verlag, 2000.
⁴Record of emergency care, Sacred Heart of Jesus Hospital, La Hormiga, Putumayo.

The child was not transferred to La Hormiga until the following day, when he was taken to the hospital at 18:00 hours. Transportation and movement in the region are difficult because the Revolutionary Armed Forces of Colombia (FARC) and the Colombian Self-defense Forces (AUC) are fighting for control of the area. According to the hospital record, a physical examination revealed the child was "in a state of shock" and exhibited obvious signs of severe dehydration, such as a depressed fontanel [note: this is the "soft spot" on the top of a baby's skull], sunken and dull eyes, dry mucus, mucocutaneous pallor, intercostal retraction, crackles in the bases of the lungs, lowered tone rhythmic cardiac noises and cardiac frequency, a depressed abdomen with no intestinal noise, rigid extremities and a profoundly depressed state of consciousness. (Here, the record refers to "Glasgow 4/5," a reference to a neurological classification of one's state of consciousness. However, this is probably not appropriate, because this illness does not appear to have been a traumatic or neurological disease.) Treatment was initiated with a saline solution administered intravenously, ampicillin, and gentamicin. Admission was at 19:30 hours and the patient died 45 minutes later.

An autopsy was performed and samples were taken of the liver, lungs and brain. They were sent to the Institute of Forensic Medicine in Bogota for analysis. As yet there have been no results.

According to the medical director of the hospital, the child’s grandfather and four-year-old brother were treated as well: the grandfather for exacerbation of chronic symptoms of severe epigastralgy (a painful burning sensation in the upper abdomen) and the brother for diarrhea and vomiting.

Although not noted in the child’s medical records, Dr. Pérez indicated the local water supply is extremely poor. Because sewage from rural homes is dumped directly into surface flows, most water sources are polluted with solid waste.

In this case, the mother apparently believes there may have been direct exposure to the sprayed mixture ("an airplane flew over") prior to appearance of the symptoms.

However, computerized spray flight records indicate that on November 14, the date of the alleged exposure, spraying took place in the townships of El Tigre, La Chorrosa, El Diamante, Campoalegre, El Venado, San Francisco and Guadualito.⁵ Guadualito is three miles south of Argelia; the other townships are further away. Although there may be slight drift of the spray mix at times, it is highly unlikely that the spray could not have drifted that far.

Even if the child were directly exposed to spraying, however, there is no indication that his illness was related to such an exposure. Even full-body exposure would have been minimal in comparison to harmful doses. Considering the dose used in aerial eradication, and assuming a conservative body weight of nine kilograms (actual weight is not on record), a corporal surface of 0,44 m2⁶ and total exposure to spraying, the dose on the cutaneous surface would be:

⁵Flight data from aerial eradication records, 2001.
⁶Corporal surface of an 11-month-old child weighing nine kilograms, according to the Boyd & West modified nomogram (a graphic representation of several lines marked to scale and arranged so that by using a straightedge to connect known values, an unknown value can be read at the point of intersection), Nelson’s Textbook of Pediatrics, 12th Edition, McGraw-Hill, 1993.

•  0. 44 m2 x 374.4 mg/m² of glyphosate (dose/ m²) = 164.73 mg
   
•  For 9 kg body weight, the total dose received would be :
   
• 164. 73 mg ¸ 9 kg = 18.30 mg/Kg
   
• Assuming a conservative absorption rate equal to 15% of the applied dose, the absorbed or "effective" dose would be:
   
• 164.73 mg x 0.15 = 24.71 mg or 2.74 mg/kg. This is 1,824 times less than the measured lethal dose⁷ for laboratory animals (5,000 mg/kg) and one and a half times less than the "Allowable Daily Intake" (ADI).⁸ Thus, the total amount of glyphosate to which the child would have been exposed even had his entire body surface been sprayed (164.73 mg) would be equivalent to 0.48 mL of the commercial glyphosate formula. Even if administered orally, this amount would be insufficient to produce the aforementioned effects.

Moreover, a review of existing literature shows no fatalities from cutaneous exposure. Mortality associated with glyphosate exposure has been observed only with the ingestion of large quantities of the commercial formula (80 to 250 mL, 184 mL on average).⁹

⁷"Lethal Dose 50," or "LD50," is a universally-applied marker for evaluating toxicity. The LD50 of a chemical is that single dose which will, on the average, result in death of 50% of a given population of experimental animals. Normally, this is a standardized test species. The oral LD50 of glyphosate is over 4,320 milligrams per kilogram of body weight, meaning that ingestion of more than 4,320 milligrams of glyphosate per kilogram of body weight would be necessary to prove lethal to the test species, in this case, rats.
⁸The "ADI" is a standard risk assessment indicator established by the World Health Organization International Programme on Chemical Safety and the U.S. Environmental Protection Agency (EPA). It indicates the maximum allowable amount of a toxic substance that can be ingested daily over a lifetime without a probability of suffering an adverse affect. Environmental Health Criteria No. 6, Risk Assessment of Chemical Substances, International Programme on Chemical Safety, World Health Organization, Genève, 1984.
⁹Talbot et al. (1991), Sawada & Nagai (1992), Tominack et al. (1993) in Environmental Health Criteria No. 159 , Glyphosate, International Programme on Chemical Safety, World Health Organization, 1994.

The child’s symptoms correspond to the unspecified signs of acute diarrhea with severe dehydration. However, those the mother reported to the municipal representative; namely, conjugated deviation of the eye and corporal instability, also suggest the presence of a major hydroelectrolytic imbalance, toxicity of a neurotoxic pesticide (a property of other pesticides, not glyphosate) or the clinical manifestations preceding incipient meningitis.

The grandfather and brother experienced similar discomforts, but no source indicates they were exposed to spraying. This could denote an infectious-contagious disease or one of toxic-food origin.

The toxic potential of glyphosate is minimal and the level of exposure does not coincide with the symptoms observed in this case. However, exposure to a more potent chemical neurotoxic agent, such as one of the organophosphate or carbamate pesticides used to protect coca (which are widely used without safety precautions both in legal and illegal agriculture in the rural regions of Colombia), could conceivably have been the cause. ¹⁰

In conclusion, a toxic etiology of the disease responsible for death cannot be identified. However, even if the child had been exposed to aerial spraying with glyphosate, the dose would not have been significant. Exposure by the cutaneous pathway is inefficient for absorption, which some authors say is around 2 percent.¹¹ The action of glyphosate is not known to produce severe diarrhea,¹² and the region’s epidemiological patterns point to the high likelihood of an infectious cause. Finally, there are no reported cases of glyphosate being responsible for human death, except in cases involving intentional ingestion of significant quantities of the undiluted commercial product. (See references on intentional exposure in footnotes 2, 3 and 10).

The samples taken during the autopsy are being analyzed at the Institute of Forensic Medicine in Bogota.

On November 22 at 19:00 hours, the second child, a four-year-old boy and neighbor of the first child, was treated at the hospital for fever and diarrhea, which began the afternoon of November 21, 2001. The patient was feverish upon admission (39ºC temperature and a cardiac frequency of 150 beats per minute). A blood test showed leukocytosis (an increase of white blood cells) with neutrophilia (the condition that results when neutrophilic cells predominate to an abnormal degree). At 06:00 hours, the patient registered "confusional syndrome, labial cyanosis, coldness, periungual cyanosis without diuresis" then "cardiorespiratory arrest. It was managed with endotracheal intubation, cardiac massage, adrenaline, atropine but without effect and was followed by death.

The final diagnosis was noted as "acute diarrhea, hydroelectrolytic imbalance and possible paraquat intoxication".¹³ According to the medical director of the hospital, the autopsy showed a massive amount of parasites migrating from the respiratory tract to the digestive tract.

¹⁰"Retrospective Study of Reported Human Health Effects From Glyphosate Exposure in Putumyo", Uribe, C. et al., June 2001.
¹¹Williams, G.M., Kroes, R., Munro, I.C., Safety Evaluation and Risk Assessment of the Herbicide Glyphosate, J Regulatory Toxicology and Pharmacology, 31: 117-165, Academic Press, 2000.
¹²Reigart, J. R., Roberts, J.R., Recognition and Management of Pesticide Poisonings, 5th Edition, See information on human exposure in notes 2 and 3.
¹³Emergency Care Record No. 49729, Sacred Heart of Jesus Hospital, La Hormiga, Putumayo.

4 .2.2 Discussion of the Findings

There was no spraying at La Hormiga on November 21, which was the first day of the acute illness observed in the second case. On November 22, the other date of possible exposure in this case, spraying took place in Brisas del Placer, Arauca, La Paloma, El Palmar, and Brisas del Guamuéz. However, only two of these locations (Palmar and Brisas del Guamuéz) are in the municipality of La Hormiga. The closest, Palmar, is located 20 miles west of La Argelia. As in the first case discussed above, this is too far for exposure to spray drift to have been a reasonable possibility.

As to the medical findings, the second child was also a case of severe diarrhea. He was ultimately found to have a definite history of massive intestinal parasite infestation and symptoms of infection (fever + leukocytosis). However, the child's parents did not claim that he had been exposed to spraying, nor was there any other evidence of exposure to aerial spraying. In fact, it is not clear why this case was brought to the attention of the Army, since it appears neither the parents nor the municipal representative made a complaint. Most likely, the hospital mentioned this case to the Army because it occurred close in time to the death of the first case discussed above (which was attributed to spraying) and in a region where aerial spraying was thought to have occurred.

As to the child’s illness and cause of death, a review of the patient’s treatment at the hospital shows the quantity and infusion rate and the sodium content of the liquids administered intravenously were insufficient to replace losses incurred earlier through diarrhea and vomiting, or to restore diuresis within a period of at least three hours and to initiate the administration of potassium, which probably was reduced substantially by diarrhea, at least during the previous 24 hours. In the end, death was the result of a severe hydroelectrolytic deficit, hypokalemia and subsequent multisystemic failure, which could not be reversed through resuscitation.

In this case, infestation with Ascaris lumbricoides, a very common intestinal parasite, is, in addition to Stongyloides stercolaris infestation, one of the most common causes of diarrhea.¹⁴ This alone could explain the origin of the disease responsible for death. Additionally, the notes of the treating physicians suggest that exposure to paraquat, a highly toxic agricultural chemical commonly used in coca cultivation in Colombia, was a possibility, although no tests for that substance were apparently conducted.

¹⁴Plata Rueda et al., El Pediatra Eficiente, Chapter 8 "Diarreas," Salvat, 1986; Nelson’s Textbook of Pediatrics, 12th Edition, McGraw-Hill, 1993.

The presence of fever accompanied by leukocytosis is the clinical event that suggests infection was responsible for the diarrhea that led to dehydration, which was the ultimate cause of death. This was aggravated by an insufficient supply of liquids and electrolytes during treatment. It is quite possible that migration of parasites to the respiratory tract and symptoms of asphyxia were contributing factors as well.

An epidemiological analysis of the cause of an outbreak of diarrhea or gastrointestinal disease must consider the time, form and place, in addition to the factors that could be responsible and the extent and intensity of exposure to them.

These infectious factors (bacterial, parasital, viral), the most common in the environment, and the toxins that produce the symptoms characteristic of diarrhea and gastrointestinal disease, as in the case of certain pesticides, must be considered when assessing the possible cause of death.

With diarrhea, the most probable cause in patients with no previous illness is infection. The risk situation in La Hormiga (described below) points to infection as a highly probable cause. Toxic agents such as organophosphate and carbamate pesticides are far more likely to cause diarrhea with fatal dehydration than is glyphosate (See footnote 13).

As to the two cases in question, the most important environmental event is water pollution resulting from solid waste disposal. This is a predominant risk factor for the people of La Hormiga and is an effective means of dissemination and exposure to infectious diarrhea.

According to health statistics, acute diarrhea (ADD) is the leading cause of death in La Hormiga.

Source: Figures on the Department of Putumayo were supplied by the National Health Institute and originate with the National Health Policy Analysis and Planning Office. SIS103, Ministry of Health.

Notes:

Due to the social and political situation in Colombia and because departmental health officials are slow about reporting data to the National Health Institute (INS), these figures could be incomplete and should be regarded with caution. Figures from recent years are not consolidated and, in some cases, none have been reported. The system was changed in 1997 and departmental officials only report when notification is mandatory, as is the case with diseases transmitted by vectors, those preventable through vaccination and sexually transmitted diseases.

All infectious skin diseases are grouped as dermatopathies, even though WHO differentiates between specific diseases such as dermatophytosis, eczemas, pruritus, allergic dermatitis, other types of dermatitis, etc. According to the INS, this change in procedure is not fully understood, especially at municipal level. As a result, there are problems with records on morbidity.

In addition, the risk of morbidity associated with water and sanitation is 57.67 percent in La Hormiga. This is one of the highest rates in the country and is near 60 points, which is the health unfeasibility limit for a community or region.¹⁵

The intensive and uncontrolled use of pesticides to protect coca crops and the potential for environmental transport of these substances after application are also factors to consider. A wide variety of highly toxic products are used in rural Colombia to protect coca against pests and blight. In most instances, for obvious reasons, these substances are not applied properly or in the correct dose, nor are they stored, handled or mixed according to safety standards. Disposal of empty containers and similar waste is usually inadequate. At some point, they could result in significant exposure to humans. Those who are most vulnerable, such as children and the elderly, would be affected first, as would certain types of wildlife. This could have been the case.¹⁶

¹⁵"Segunda Fase de Actualización del Inventario Nacional de la Calidad del Agua," Colombian Ministry of Health, 1998.
¹⁶op. cit. in Note 11.

The following table shows the products used most often, their toxicity and their measured lethal dose.

Source: "Retrospective Study of Reported Human Health Effects From Glyphosate Exposure in Putumyo", Uribe, C. et al., June 2001.

There is no evidence in the two cases discussed above of either exposure to aerial spraying or of glyphosate as a causal agent in the deaths of the two children in La Hormiga.

APPENDICES

CLINICAL HISTORY NO. 1

CLINICAL HISTORY NO. 2

LETTER: NATIONAL INSTITUTE OF LEGAL MEDICINE AND FORENSIC SCIENCES

REPORT:  ASSESSMENT OF A PATIENT IN THE TOWN OF SAN BERNARDO (NARIÑO, COLOMBIA), Bogota, September 21, 2001