Pituitary dysfunction in survivors of Russell's viper snake bite envenomation: A prospective study
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.241378
Source of Support: None, Conflict of Interest: None
Keywords: Empty sella, pituitary dysfunction, Russell's viper snakebite
Morbidities following snake envenomation (SE) are caused by local wound infection, gangrene, and renal insufficiency, especially with vasculotoxic bites. Neurotoxic snakebite can cause ptosis, ophthalmoplegia, bulbar weakness, respiratory muscle involvement, and limb weakness. Elapids usually cause neuroparalytic manifestations, although coagulopathy may also occur rarely.
One of the rarely reported sequelae of SE is hypopituitarism, which exclusively occurs with vasculotoxic SE. The prevalence of this complication has not been systemically studied. It is surprising that even though SE is seen over a large geographic distribution in India, reports of hypopituitarism are sparse.,,, This could be due to the under recognition, lack of awareness, and underreporting of this entity.
There are more than 20 case reports of endocrinal insufficiency in the world literature and all were due to the occurrence of vasculotoxic snakebite. The previous cases reported were mostly from southeast Asian countries and there was one case reported from Brazil. There are only two prospective studies dealing with this subject, although they are fraught with a selection bias., The various predictors of hypopituitarism mentioned in the previous studies include the presence of renal insufficiency and the severity of envenomation. In the present prospective study, we evaluated the prevalence of pituitary dysfunction at the baseline and 6 months of follow-up in patients who had undergone SE. We also attempted to determine the predictors of hypopituitarism following SE.
This prospective study was conducted in the Department of Endocrinology and Emergency Medicine, Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh, India. Few cases were also collected from Shimla (JM) and Madurai (SS). All cases with an alleged history of snakebite with the clinical picture suggestive of envenomation by a poisonous snake and who received anti-snake venom were included. The study was conducted from July 2013 to August 2014. The patients were excluded if hydrocortisone infusion had been received along with the anti-snake venom prior to taking blood samples for the hormonal analysis. The study was approved by the Intramural Institute Ethics Committee, PGIMER, Chandigarh. Random blood samples were taken as soon as the patients reported to the emergency department.
The snake was identified based on the appearance of the live or dead snake brought along with patients; the description of the snake by the patients or bystanders at the time of bite; the matching of the description with the representative photographs of the locally prevalent snakes; and, by observing the photograph of the snake taken by a bystander after it was killed. The snake type was reconfirmed by an expert toxicologist (SS, A. Bhalla, NS). Patients who had suffered from a snakebite in whom the snake type could not be identified were categorized as the ‘not identified (NI)” group. The patient samples for all biochemical, hematological and coagulation profiles, arterial blood gas analysis, and hormonal profile were evaluated. The clinical assessment of vital parameters, urine output, and renal functions at baseline was also recorded. The intensity of the snake bite was analyzed by applying the scoring systems including the local effects, systemic features, and coagulation abnormalities. The clinical examination and hormonal profile were repeated at a follow-up of 6 months. The hormonal profile included the assessment of the growth hormone (GH), insulin like growth factor-1 (IGF-1), thyroxine (T)3, T4, thyroid stimulating hormone (TSH), cortisol, adrenocorticotropic hormone (ACTH), as well as luteinizing hormone (LH), follicular stimulating hormone (FSH), and estrogen in female, and testosterone in male patients. All hormones were analyzed by the electro-chemiluminescence method (COBAS 800, Roche Hitachi Diagnostics, Germany). IGF-1 was measured by DiaSorin, Liaison, Germany, with a lower detection limit of 25 ng/ml. The sera were stored at −80°C, and all samples were run by a single assay to avoid variability.
Hypopituitarism was defined as deficiency of one or more of the pituitary hormones. Secondary hypothyroidism was defined as a low thyroxine (T4) level with a low or inappropriately normal thyroid stimulating hormone (TSH). Plasma cortisol of less than 350 nmol/L (established from our own in-house assay) with inappropriately low ACTH was taken as glucocorticoid insufficiency. Growth hormone deficiency in adults was defined as low-age and gender-specific IGF-1 in the presence of at least one pituitary hormone deficiency. Patients with low IGF-1 and/or glucocorticoid were subjected to insulin-induced hypoglycemia and/or ACTH stimulation test to examine the GH and cortisol reserve. A peak GH of less than 5 ng/ml was considered as being representative of GH deficiency. Hypogonadism in male patients was defined as the presence of serum testosterone level of less than 9 nmol/L with an inappropriately low or low normal LH and FSH level, with or without clinical symptoms. In female patients, it was defined as the presence of low or inappropriately normal FSH or LH level or the presence of secondary amenorrhea. This was further corroborated by a low estradiol level that was assessed following SE. Hyperprolactinemia was defined as serum prolactin of more than 20 ng/ml in male, and more than 25 ng/ml in female patients; and, hypoprolactinemia was defined as the presence of serum prolactin of less than 5 ng/ml in either gender. To determine central diabetes insipidus, the urine and plasma osmolality were assessed by the freezing point depression method (Fiske, USA). Those patients with confirmed hypopituitarism were subjected to magnetic resonance imaging (MRI) of the hypothalamo–pituitary axis. The severity of illness was assessed by various tools including the Glasgow coma scale (GCS) score, the acute physiology and chronic health (APACHE II) score, the sequential organ failure assessment (SOFA) score, and the snakebite severity score (SBSS).,,
Quantitative data was presented as mean ± standard deviation [SD], or as median + interquartile range, and the qualitative data were presented as number and percentage. Categorical data was compared using chi-square and Fisher's exact test. Normality of quantitative data was checked by the Kolmogorov–Smirnov test. Independent t-test was used for normally distributed data and Mann–Whitney U-test for skewed data. To determine the predictors of hypopituitarism, logistic regression analysis was done. All statistical tests were two-sided and were performed at a significance level of P = 0.05.
Of the 76 patients (44 female), 16 patients were lost to follow-up, and a repeat hormonal profile could be performed in 60 patients (41 with vasculotoxic SE) at the end of the study with a majority of the susceptible population in the middle-age group (mean age, 37.6 ± 14.9 years). The mean lag period at presentation was 32 ± 20 h. Forty-one patients had local complications in the form of cellulitis, swelling, hemorrhagic blister, and tissue necrosis of the limbs [Figure 1]a-c]. Systemic manifestations in the form of vomiting, headache, ptosis, respiratory distress, and bleeding tendencies such as hematuria, venipuncture site bleed, and hematemesis were noted in 42 (55.2%) patients. Thirty-five patients (46.1%) had coagulopathy, 20 patients (26.3%) had acute kidney injury (AKI), and 8 of 76 patients (10.5%) needed renal replacement therapy (RRT) in the form of hemodialysis.
Hematemesis, bleeding, AKI, and requirement for dialysis occurred only in viper bite. The mean scores of GCS, APACHE II, SOFA, and SBSS at presentation were 14.71 ± 1.56, 4.86 ± 5, 1.17 ± 2.17, 5.37 ± 2.94, respectively. Local complications in the form of limb cellulitis were noted in the viper (92.6%) and NI (presumed viper bite, 85.7%) categories [Table 1]. Systemic manifestations in the form of hematuria were noted in 8 patients (10.5%), bleeding from orifices/bite site in 3 patients (3.9%), and hematemesis in 2 patients (2.9%).
Of the 60 patients, majority of the patients with vasculotoxic snakebite were from Himachal Pradesh region. Six patients developed acute asymptomatic anterior hypopituitarism. Of the 6 patients, 4 had envenomation caused by Russell viper bite, and in 2 patients, the nature of the snake could not be identified; however, according to the clinical profile, they had envenomation by the Russell's viper. The most common deficiencies were low growth hormone in 5 patients (83%) and glucocorticoid deficiency in 5 (three with secondary adrenocorticotrophic [AI] and 2 with primary AI deficiency) patients. Gonadotropin deficiency was present in 3 (50%) patients. One male patient had a high gonadotropin level with normal testosterone level. Hypothyroidism was present in only 1 patient. None had central diabetes insipidus. Two patients each had two, three, and four hormonal deficiencies, respectively [Table 2]. At the 6-month follow-up, there was no hormonal deficiency in any of the 54 patients who were euhormonal at the baseline. In the remaining 6 patients, persistent GH deficiency was present in 4 (with a low IGF-1 and a poor GH reserve), gonadotropin deficiency in 3, secondary AI in 2, and primary AI in 1 patient (confirmed by the provocation test). One patient each had secondary hypothyroidism and a low prolactin level but none had diabetes insipidus (DI) [Table 1] and [Table 2]. At a 6-month follow-up, contrast-enhanced MRI showed an empty sella [Figure 2] and [Figure 3] in 2 patients, and 4 patients had a normal imaging [Figure 4]a and b]. Two patients with empty sella had four hormonal deficiencies (GH, cortisol, gonadotropin, and thyroid/prolactin). Only 3 patients with chronic hypopituitarism were symptomatic. The presence of renal insufficiency, renal replacement therapy, coagulopathy, thrombocytopenia, SBSS, and other tools could not predict occurrence of hypopituitarism. Only viper bite showed a positive correlation hypopituitarism but this correlation could not reach statistical significance (P = 0.09).
In the 16 patients who were lost to follow-up, baseline evaluation of hormones was done [Figure 5]. One patient had hypothyroidism along with low estradiol and high FSH levels consistent with her post-menopausal status. Seven patients had excess gonadotropins levels consistent with their post-menopausal status. Four patients had a high cortisol level with normal ACTH level, and 3 patients had a low cortisol level with low ACTH level consistent with secondary adrenal insufficiency. Five patients had a low IGF-1 level at the baseline; however, further assessment of GH/IGF-1 axis could not be done as they were lost to follow-up. In these patients also, the logistic regression analysis did not show any statistically significant correlation of different variables with the incidence of hypopituitarism. Only the type of snakebite (viper) showed some correlation with hormonal insufficiency (P = 0.09).
SE-induced hypopituitarism may occur due to a vasculotoxic (Russell viper) bite only in a small but significant proportion of patients. Hypopituitarism is found in 10% of the patients. Acute hypopituitarism generally persisted on retesting at 6 months. Contrary to the popular belief, it is prevalent in north India also. There is no predictor of occurrence of hypopituitarism. Acute hypopituitarism may be completely asymptomatic. MRI abnormality of the sella is not universal in patients with chronic hypopituitarism.
Snake bite is a major health hazard in tropical countries, often leading to significant morbidity and mortality among developing countries. Chronic complications related to snakebite are acute kidney injury followed by chronic kidney disease, bleeding manifestations in vital organs leading to sequelae, and hypopituitarism, which was the most underrecognized and underreported complication. In the present study, overall, 10% of the patients (20%, if only patients with viper bite are considered) had acute hypopituitarism, which continued to persist at 6 months of follow-up, leading to chronic hypopituitarism.
In the current prospective study, out of the 6 patients who developed hypopituitarism, 4 patients had definite viper bite and two had probable viper bite (all the patients were from Himachal Pradesh). Five patients (83.3%) who developed hypopituitarism had local complications in the form of cellulitis of limbs but systemic manifestations were noticed only in 2 patients (33.2%) in the form of hematuria, headache, and vomiting. Occurrence of hypopituitarism was not predicted by the presence of acute kidney injury, requirement of hemodialysis, coagulopathy, disseminated intravascular coagulation (DIC), APACHE-II score, GCS score, SOFA score, snakebite severity score, and local cellulitis. Previous studies by Tun-Pe et al., and Golay et al., demonstrated the occurrence of acute kidney injury as a predictor of hypopituitarism., The first study was retrospective in nature, and in the second study, there was a selection bias as only those patients who developed renal insufficiency were included. In the study by Rajagopala et al., thrombocytopenia, massive bleeding, and the need for transfusion were the predictors of acute or chronic hypopituitarism. We could not find any such correlation in our study. It may also be noted that the limited size of the study population most likely hampered the identification of predictors of pituitary dysfunction.
Though viper snakes are widely distributed all over the world, the occurrence of hypopituitarism following envenomation is reported mostly from southeast Asian countries, though the first case of hypopituitarism following Bothrops (Pit viper) bite was reported from Brazil. This could be probably due to an abnormal pituitary morphology, the venom variability, and increased recognition of the association of hypopituitarism with the viper bite due to the presence of previously published literature. The pathogenesis of pituitary damage leading to hypopituitarism could be a triple hit process; in the first stage, there is pituitary stimulation and enlargement caused by the direct effect of the venom or an increased capillary permeability; in the second stage, major bleeding-related ischemia and DIC-related micro-thrombosis leads to ischemic necrosis; rarely, internal carotid artery bleed may lead to a raised intracranial pressure and pituitary stalk compression; and in the third stage, chronic antigen leakage may lead to ongoing pituitary destruction, as described by Goswami et al., in patients with Sheehan's syndrome.,, In our study, hypopituitarism developed in the absence of major intracranial bleeds and hypotension; and, all the patients with chronic hypopituitarism had no added hormonal deficiencies when they were compared with patients who developed acute hypopituitarism; therefore, the third mechanism is still conjectural. Similarly, if only coagulopathy and DIC are considered as being responsible for this process, hypopituitarism should have been documented in patients with elapidae envenomation in the past.
Hypopituitarism during an acute presentation predominantly showed involvement of the growth hormone, corticotropin, gonadotropin, thyroid hormone, and prolactin, in that order. During the follow-up, the same pattern of hormonal deficiency persisted. However, in 2 patients who had secondary adrenal insufficiency at presentation, at follow-up visit, the ACTH level recovered but the cortisol level was still persistently low. We did not have a CT scan of the adrenal glands to support the presence of superimposed primary adrenal insufficiency. It is usually believed that chronic hypopituitarism develops de novo after recovery from acute envenomation. Similar to the findings of our study, acute pituitary insufficiency appeared to be persistent in 11 of the 12 patients over 8–156 weeks of follow-up in a study by Proby et al. Chronic hypopituitarism can be totally asymptomatic; therefore, basal and dynamic pituitary function tests, including insulin-induced hypoglycemia and ACTH stimulation test, should be performed in all the affected patients, if indicated at 6 months of follow-up, as suggested by Tun-pe et al.
In our study, none of the patients had central diabetes insipidus. Posterior pituitary insufficiency is rarely documented following a snakebite; only 1 out of 1000 patients in a large series by Eapen et al., had central diabetes insipidus and so far only 4 cases having posterior pituitary insufficiency have been reported in the English literature.,
Acute hypopituitarism other than glucocorticoid deficiency is difficult to diagnose in any critically ill patient as sickness per se may lead to widespread alteration of the growth hormone, gonadotropin, and thyrotropin axes. Even the glucocorticoid threshold required to diagnose hormonal deficiency in critically ill patients in the intensive care unit setting is debatable. The glucocorticoid measurement in the previous studies related to snakebite were associated with assay interference as the patients were on hydrocortisone or prednisolone. In our study, all the samples were obtained before intravenous hydrocortisone had been given to prevent polyvalent anti-snake venom (ASV)-related anaphylactoid reaction. Though hypotension and hypoglycemia are red flags that interfere with the diagnosis of acute glucocorticoid deficiency, they are not universally present; hypotension in the setting of snakebite could be multifactorial, for instance, due to an increased capillary permeability leading to relative hypovolemia, venom-induced activation of bradykinin and kininogen bleeding, thromboembolism, and anti-snake venom related anaphylactoid reaction.
There is a wide range of lag period between the envenomation and the clinical diagnosis of hypopituitarism. In a Burmese study, hypopituitarism was diagnosed as late as at 4 years, and in a report from south India, the lag period was recorded as being up to 20 years. We illustrated one such patient from Madurai (Tamil Nadu) who presented after 10 years of envenomation with hyponatremic encephalopathy caused by glucocorticoid and thyroid hormone insufficiency. Another patient from Chamba Valley (Himachal Pradesh) presented after 4 years with secondary amenorrhea, and on evaluation, was found to be having panhypopituitarism. Though hypopituitarism may present after a variable lag period, we presume that it occurs usually within the first few days of envenomation. Therefore, we have not reevaluated the non-hypo-pituitaric patients beyond the period of 6 months. In our prospective study, in all of the patients in whom chronic hypopituitarism manifested, the presence of chronic hypopituitarism occurred in continuation with acute hypopituitarism.
Snakebite-related hypopituitarism reported so far in the literature occurred mostly among the young population probably due to their greater outdoor activity. Similar findings have been observed in our study, though there was no gender predilection [Table 3].,,,,,,,,,,,
The limitations of our study were the recruitment of a small number of patients with vasculotoxic snakebite, unavailability of MRI at the baseline evaluation, and the lack of CT scan images of the adrenal gland in the 2 patients who developed primary adrenal insufficiency similar to the manifestation seen in the Waterhouse–Friderichsen syndrome. The identification method of using photographs of snakes in some of the patients is a less reliable method. Identification of venom antigen is the most definite method of categorizing the nature of snakebite, as previously highlighted by Warrel. Random blood sample was taken for the analysis of cortisol and testosterone at the baseline evaluation. However, both these hormonal levels follow a strict circadian variation. This variability in the levels was taken care of by evaluating a basal morning sample for the hormonal analysis on the follow-up visit. The sensitivity of IGF-1 for establishing the diagnosis of adult GH deficiency is very low. Limiting GH testing to those patients with low IGF-1 levels would likely lead to under-recognition of GH deficiency. However, we believe that, in a well-nourished person without any underlying illness, a normal IGF-1 level usually excludes the presence of GH deficiency.
Acute asymptomatic, and chronic symptomatic or asymptomatic hypopituitarism are important sequelae of vasculotoxic snakebite in a significant proportion of patients. Routine pituitary hormonal screening should be done in all patients after 6 months of envenomation. Chronic pituitary dysfunction can occur even in the presence of normal pituitary imaging.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3]