Vultures. The Diclofenac Epidemic

Another major problem, probably the most serious for vultures in history and prehistory concerns the diclofenac issue (Woodford et al. 2008). Van Dooren (2011) notes that today, six of the 23 recognized vulture species are threatened or critically endangered on the ICUN Red List of Threatened Species, with the largest declines being of five species in India. This event has been termed the 'global vulture crises' (Olea and Mateo Thomas 2009). Some vultures in India have declined by as much as 90-99% of their previous numbers, approaching local, regional and possibly even global extinction. Prior to this catastrophic event, Griffon vultures were very common throughout South and South-East Asia. Seven species of vultures are resident in South Asia, and three of these seven—the White-rumped Vulture, Slender-billed vulture and Red-headed vulture—are also resident of South-East Asia (Houston 1985; Clements et al. 2012).The White-rumped vulture was considered one of the most abundant large birds of prey in the world. Vulture populations had declined marginally in much of Asia in the first half of the 20th century, but were still common on the Indian subcontinent, largely supported by a large supply of livestock carcasses. Records of the sudden deaths of vultures were first made in India, in studies of Gyps vultures in Keoladeo National Park, Rajasthan, India (Prakash 1999; Gilbert et al. 2004).

This study was supported by later research across India (Prakash and Rahmani 1999; Prakash et al. 2003) and Nepal and Pakistan (Virani et al. 2001; Gilbert et al. 2002). In the late 1990s, it was evident to observers that the Indian populations of White-rumped vulture, Indian vulture and Slenderbilled vulture had totally crashed, with similar declines also observed in Nepal and Pakistan. The decline of the vultures was also noted in the media (BBC 2004; PIB 2005; Gill 2009).

At first three species were affected, the White-rumped vulture, Longbilled vulture and Slender-billed vulture. The decline, which took place between 1993 and 2000, led to the classification of these vultures in 2001 as Critically Endangered (BirdLife International 2001). Over 15 years (1992 to 2007) the population of the White-rumped vulture was one thousandth of the previous population, that of the Red-headed vulture was less one tenth and that of the Slender-billed vulture was reduced to about 1,000 birds (Cuthbert et al. 2006; Prakash et al. 2007).

However, at that time Gilbert et al. (2004) argued that 'attempts to quantify the Asian vulture decline have been complicated by the lack of pre-decline data on population size in these formerly abundant species'. A noted exception was the study by Galushin 1971, and that of Prakash 1999. Nevertheless, the evidence until at least until 2008, was that the populations were still in decline (Green et al. 2004; Gilbert et al. 2006). The White-rumped vulture, formerly possibly the commonest vulture in Asia was particularly affected, as it declined by 99.9% since 1992 (Prakash et al. 2007). Non Gyps vultures were also affected, as the Egyptian vulture and Red-headed vulture were classified as Endangered and Critically Endangered respectively (Cuthbert et al. 2006; BirdLife International 2008).

The deaths of millions of vultures on the Indian subcontinent had a marked impact on the natural ecosystems and human quality of life. Facultative scavengers, mostly feral dogs (Canis lupus familiaris) greatly increased, but they were unable to eliminate carcasses with the same efficiency as the vultures. Large numbers of decaying carcasses littered the landscape contributing to diseases. Large numbers of dogs were attracted to the carcasses, leading to the incidence of rabies that affected thousands of people (Swan et al. 2006; Gill 2009). Gross (2006) points out that if rats increase, bubonic plague and other rodent-transmitted diseases may also increase. For the sky burials, in which Zoroastrian Parsis and Tibetans left their dead on platforms for vultures to eat, in the past hundreds of vultures could clean a corpse in less than a hour. With local extinction of vultures, alternatives had to be found.

Before the diclofenac identification was made, Pain et al. (2003) investigated all the possible causes of the population decline. This study noted that during the 1970s and 1980s, Gyps vultures were rare or absent in much of Southeast Asia, but G. bengalensis and also G. indicus, were very common in India, especially in and near urban centers. Their abundance was principally because of the religious prohibition of killing cattle in northern and central states of India. This resulted in the dumping of large numbers of cattle carcasses in rural landscapes (Grubh et al. 1990). The vulture populations grew so dense, they were an aircraft risk (Grubh et al. 1990). In Southeast Asia (Cambodia, Laos, Vietnam, Malaysia, Thailand, and even Yunnan Province, China), the vultures were much rarer (see also Timmins and Ou Ratanak 2001). Several reasons were advanced for the low populations and/or declines in southeast Asia.

Hunting was blamed for the decimation of wild ungulates that provided vultures with food (Srikosamatara and Suteethorn 1995; Duckworth et al. 1999; Hilton-Taylor 2000). In addition, changes in husbandry of domestic stock contributed to the reduction of food sources (Cambodian Wetland Team 2001). 'It seems likely that food supplies are no longer predictable enough to allow regular breeding' (Pain et al. 2003: 661).

Habitat loss was seen as less important, except for a minor effect on the wild ungulates (see also Thewlis et al. 1998). Vultures may also have been killed by people, when the vultures were gorged with food on the ground (Thewlis et al. 1998). Pain et al. (2003: 661) wrote that 'The role of agrochemicals remains unclear; there is no persuasive indication that they can explain region-wide losses in Southeast Asia, although they may have caused local declines' (see for example, the dated example of Cheke 1972). The role of infectious diseases was also ruled out, or as 'circumstantial evidence.' The decline of raptors, which was extreme in Laos, also affected other scavenging birds, e.g., such as the Greater Adjutant (Leptoptilos dubius, Gmelin 1789), Black kite (Milvus migrans Boddaert, 1783), Brahminy kite (Haliastur indus Boddaert, 1783), and Large-billed crow (Corvus macrorhynchos Wagler, 1827) (Lekagul and Round 1991; Thewlis et al. 1998; Wells 1999; Duckworth et al. 1999, 2002; Round 2000).

Laos, with some of most extreme declines, also had large areas of suitable habitat for vultures, with low human population densities, and has little evidence of environment—contaminating chemicals. The decline in open-country wild ungulates during the late 20th century from hunting and changes in livestock husbandry were the only known factors for vulture decline (Duckworth et al. 1999). Cambodia, with Gyps populations had large areas of open landscape, sparse settlements, wild ungulate populations and extensive free-ranging of domestic cattle (Timmins and Ou Ratanak 2001). Therefore, despite the lack of solid evidence for the reasons for the vulture declines in Southeast Asia, one initial explanation was that 'food shortage appears to be the most credible general explanation, although other factors including persecution and contaminants may have played a part locally' (Pain et al. 2003: 662).

India was differentiated from Southeast Asia, because of the high provision of cattle carcasses. Nevertheless, there were drastic declines of G. bengalensis and also G. indicus in India, for example in the study area of Keoladeo National Park where declines were about 95% between the mid- 1980s and late 1990s (Pain et al. 2003; see also Prakash 1999; Prakash et al. 2003). The reported reductions were the result of very high mortality of both adults and juveniles. Sick birds were described as having a slumped posture after which death occurred. At this time, the cause was unknown, as Pain et al. (2003: 664) report the 'The population decline and high mortality was unexplained' even though there was no alteration of the food supply.

The Bombay Natural History Society conducted surveys in 2000, on the model of earlier surveys in 1991-1993 covering states in north, west, and east India. The lowest declines of the Gyps vultures were of G. bengalensis and G. indicus; 96% and 92%, respectively. Note that at the time G. indicus was a combination of two currently recognized species, the Indian vulture (G. indicus) and the Slender-billed vulture (G. tenuirostris) (Rasmussen and Parry 2000). Other scavengers did not decline, and feral dogs increased in population over India (Cunningham et al. 2001). There was also no evidence that starvation was a factor for vulture deaths in both India and Pakistan (Gilbert et al. 2002; Prakash et al. 2003).

Pesticides were also investigated as possible factors for the collapsing populations. Although large amounts of pesticides are utilized in India, no chemical or pesticide was identified that could have impacted vultures so severely. For this to be the case, it would have to be applied across a vast area or used in a new way that increased its contact with vultures within the previous 10-20 years. Numerous toxicological analyses were conducted of vulture carcasses from Pakistan, testing for organochlorines, organophosphates, carbamates, and heavy metals; however none were found at toxic levels (Oaks et al. 2001). There were also no regional patterns and declines in other species. These suggested that contaminant poisoning could not be the sole factor (Pain et al. 2003).

The hypotheses were that the declines were the result of 'a simultaneous subcontinent-wide exposure to a toxic contaminant or a rapid spread of disease through the Gyps vulture population'; with the 'international, cross border nature of the problem hinting at the latter possibility' (ibid. 665). The extinction of colonies of vultures was also seen as possibly due to either vulture mortality or desertion by survivors when the population declined below a certain level. Nevertheless, at this point, prior to 2004, some studies of dead birds began to notice renal and visceral gout (crystallization of uric acid in the tissues) in most of them and enteritis in many of the Indian birds (Pain et al. 2002) and G. bengalensis from Pakistan (Oaks et al. 2001; Gilbert et al. 2002).

Visceral gout in most birds indicated similarity in the death factor. As the gout was acute, appearing a few hours before the bird died, it was hypothesized to be the result of another disease rather than the main disease (Pain et al. 2003). This evidence at the time was seen as indicating an infectious disease. 'Although we cannot be certain that an infectious disease is responsible until a causal agent has been identified, this is currently the most tenable hypothesis, so it is important to consider the implications of this explanation' (Pain et al. 2003: 666).

This infectious disease could have spread to all the eight Gyps species due to the social nature of the Gyps species in breeding, roosting and feeding, the very wide foraging ranges, the lack of geographical isolation of any Gyps species, and migratory behavior between Europe, Africa and Asia, sometimes through areas where the disease is prevalent (Houston 1974, 1983; del Hoyo et al. 1994; Griesinger 1996; Ferguson Lees and Christie 2000; Prakash et al. 2003). The spread could be from the three affected Gyps species (G. bengalensis, G. indicus, G. tenuirostris), through Gyps fulvus and Gyps himalayensis. Pain et al. (2003: 666) gave a detailed list of potential routes which we describe below.

The first routes of disease transmission would be (A) west through southern Iran into the Zagros mountains (G. fulvus); (B) northwest from Afghanistan and northern Iran to the Caucasus (G. fulvus); (C) north through the Pamir Knot and to the Tien-Shan of the former Soviet Union (G. fulvus and G. himalayensis); and (D) northeast from the Himalayas onto the Tibetan Plateau (G. himalayensis). A fifth route (E) would be across the Strait of Hormuz from southern Iran to the United Arab Emirates (G. fulvus).

The second stage of disease spread, through G. fulvus, would extend (F) into the southern Alps and Pyrenees in Europe and (G) through Middle Eastern mountains and thence to Ethiopia and sub-Saharan Africa. It would also enter Africa from Jordan and Israel into Egypt and into sub-Saharan Africa, or pass through across Saudi Arabia and Yemen and to Djibouti. Any gaps between Gyps species distributions would be overcome by species mixing, the result of the extreme broad foraging and migratory ranges of the sub-species, up to 1000 km (Houston 1974, 1976; Ferguson Lees and Christie 2000). However, at the time of the study by Pain et al. (2003: 666), concerning Gyps fulvus it was 'not yet clear, however, whether they are susceptible to the mortality factor affecting other Gyps vultures.' The disease hypothesis already seemed a little weak, as it did not seem to contribute to mortality of Gyps vultures in other parts of the world (Benson 2000).

Soon after this study by Pain et al. (2003), new ground was broken and the 'infectious disease' was identified. It was not an infectious disease. Gross (2006) records that in 2004 scientists from the United States-based Peregrine Fund found the cause. The role of diclofenac was discerned, despite other possibilities such as reductions in food availability and pesticide poisoning which had also killed many vultures in the past. In the mid-1990s, livestock farmers in India and later in Pakistan and Nepal, began treating their cattle and water buffaloes with the non-steroidal anti-inflammatory drug (NSAID) diclofenac.

Diclofenac, a non-steroidal anti-inflammatory drug, apparently was so toxic to vultures that the deaths could have been due to a miniscule proportion (between 1:130 and 1:760) of livestock carcasses, which exist at carcass dumps, the traditional method of livestock disposal in South Asia (Green et al. 2004; Oaks et al. 2004; Gross 2006). Diclofenac was used widely on the Indian subcontinent from the early 1990s (Green et al. 2004; Oaks et al. 2004). This coincided with the period of vulture deaths. Evidence emerged that vultures and some other scavenging birds were fatally affected after eating the flesh from an animal that died after recent treatment with a veterinary dose of diclofenac (Green et al. 2006; Cuthbert et al. 2007; Green et al. 2007). This would be the case even if a very small proportion of the livestock carcasses had been treated with the drug (Green et al. 2004).

A study by Naidoo (2007) found visceral gout in dead vultures in India, Nepal and Pakistan (see also Cunningham et al. 2001, 2003; Virani et al. 2001; Oaks 2001; Gilbert et al. 2002). Research found that the visceral gout was caused by diclofenac (Oaks et al. 2004). However, visceral gout may occur after several types of infectious and non-infectious diseases that contribute to renal dysfunction. The main medical effect on the birds was renal failure and visceral gout. Vultures do not have an enzyme to break down diclofenac. Hence, they died within 36-58 hours of ingesting toxic doses, which were actually normal veterinary doses in cattle. This disease occurs when kidney failure—severe renal dysfunction—causes a buildup of urates in the internal organs, with resultant anorexia and emaciation. The reduction in renal filtration increases blood uric acid levels (Lumeij 1994). Uric acid precipitates form a chalky-white coating over the visceral organs, especially the liver and heart and this may be deposited in the body tissues.

In 2006, the Drug Controller General of India issued a notification to all the State drug controllers to withdraw the licences for the manufacture of diclofenac within three months (May 11 to August 11, 2006). After the ban on the use of diclofenac, the populations of the Slender-billed vulture in Pakistan increased by up to 52% in 2008 (Chaudhry et al. 2012). In the study, the largest known breeding colony of Slender-billed vultures in Pakistan was monitored before the ban from 2003 to 2006 and also from 2007 to 2012 after the ban. The number of vultures were recorded to have declined by 61% before the ban and increased by 55% after the ban. Similar increases have been reported for Slender-billed vultures in India and White-rumped vultures in Nepal. The report notes that the numbers may grow to the level before the diclofenac epidemic on condition that the adult mortality remains at a low level, more birds are attracted to nesting and there is sufficient food.

In this study, the evidence showed that after the diclofenac ban, the breeding population regained the numbers of 2003, but the nest occupancy and productivity of the birds did not recover as fast. The authors speculate that there was a change in the age structure of the population between 2003 and 2012; the ratio of breeding adults to subadults/juveniles changed from 1.8 to 1.2 in 2012. Therefore, the lower nest occupancy results from the higher proportion of subadults and juveniles. Possibly the adults were more affected because they accessed more of the poisoned meat than the younger birds and hence died in larger numbers (Chaudhry et al. 2012).

These authors also acknowledge that local livestock producers possibly changed carcass disposal methods after the vulture population decline, e.g., using burial or burning as disposal methods. If these methods continued during the recovery of the vulture population, the lowered food supply would result in the lowered habitat carrying capacity.

African vultures have also declined, mostly due to factors other than diclofenac. As pointed out by Swan (2004: 205) 'To date, only diclofenac has been identified as a risk for vultures India and Pakistan..., but diclofenac, as well as other NSAIDs, pose a danger to Gyps vultures across their geographic range. Swan (2004: 205) further notes 'it is not yet known if diclofenac is similarly toxic to other Gyps species (i.e., G. africanus, G. ruppellii, G. coprotheres, G. fulvus and G. himalayensis). The African White- backed vulture G. africanus is considered the nearest relative of the Oriental White-backed vulture (G. bengalensis), thus potentially the most likely to be similarly affected.'

Botha et al. (2012) describe African vultures 'as under severe pressure from a range of factors and populations of some species have been in drastic decline over the last 30 years.' They acknowledge the factors for decline in African vultures are more complicated than those for Asian vultures. In West Africa some species have declined by as much as 85% and there is an average decline of about 42% (see also Rondeau and Thiollay 2004). In East Africa as well, the Lappet-faced vulture, Egyptian vulture, African White-backed vulture and Hooded vulture have seriously declined (Ogada and Buij 2011; Virani et al. 2011). However, vulture populations in southern Africa are comparatively stable, at least over the last 20 years (Monadjem et al. 2003). Problems indentified on the horizon include the increased demand for vulture body parts for the juju trade, and veterinary medicines that are lethal to African vultures (Naidoo et al. 2009). However, a major problem with the African context is the lack of monitoring (Anderson 2004). There are few qualified observers, limited funding and logistical challenges (Botha et al. 2012).

An important finding was that Turkey vultures are not affected by diclofenac (Rattner et al. 2008). A toxicological study involving oral applications of diclofenac to selected Turkey vultures found no toxicity, visceral gout, renal necrosis, or elevate plasma uric acid, even at concentrations more than 100 times the estimated average dose that kills Gyps vultures. When the Turkey vultures received 8 or 25 mg/kg of diclofenac, the authors estimated the plasma half-life of diclofenac to be 6 hours, as examination showed it cleared from their systems with no residues in the liver or kidneys. The study recommended further studies of this important finding.