A Simple Explanation for Intractable Pruritis
Was this patient's pruritus the manifestation of a single multi-system disease involving the lungs and nervous system, or a separate entity occurring in the presence of other unrelated medical problems?
A 71 year old male of Honduran descent was referred to the Einstein Allergy/Immunology clinic because of severe generalized pruritus without a rash of 9 months duration. He had been previously evaluated by several specialists and had been treated with antihistamines and oral corticosteroids without success.
Pruritus, or itch, is defined as an unpleasant sensation that provokes the desire to scratch. Despite its apparent triviality, it can be terribly troubling for patients, especially since it is so persistent, prone to occur completely unexpectedly and so difficult to treat successfully. There is a long list of dermatological and systemic medical conditions that can cause pruritus without rash and successful treatment usually depends not solely on drug therapy but requires intensive medical investigation to discover the cause of the problem. This is often difficult to accomplish so that this patient’s history of 9 months of pruritus despite medical treatment and numerous specialist evaluations is not uncommon.
The patient's medical history included a number of other medical problems. Three years prior to the visit he had developed gradually progressive, bilateral upper and lower extremity numbness and weakness that was diagnosed as chronic inflammatory demyelinating polyneuropathy (CIDP) by electromyography and nerve conduction velocity (EMG/NCV) testing. He was placed on oral corticosteroids for the CIDP, although it is not clear that this treatment was at all helpful. He reported having had a “stroke” nine years prior although repeated imaging of the brain demonstrated no evidence of an old infarct. He was also diagnosed with asthma and chronic rhinosinusitis 3 years before and had hyperlipidemia, type 2 diabetes mellitus, and benign prostatic hyperplasia.
There were no obvious environmental factors in his home such as pets, feathers or dampness. He was not a cigarette smoker. Travel history was notable for emigration to the United States 35 years previously, and travel abroad the most recent being a trip back to Honduras 3 years previously. Medications at the time of the first visit included hydrochlorothiazide, valsartan, atorvastation, aspirin, metformin, albuterol, ipratropium, salmeterol/fluticasone, and tamsulosin.
At this point, this patient presents a common medical dilemma; is his pruritus the manifestation of a single multi-system disease involving the lungs and nervous system, or a separate entity occurring in the presence of other unrelated medical problems? Traditionally, when faced with complex medical problems, physicians are taught to apply Occam’s razor (the law of parsimony); that the simplest (usually single) diagnosis is usually the correct one. However, as patients age and develop multiple medical problems, it is not unusual to find that a single diagnosis cannot explain every symptom or sign. This may be especially true with a symptom such as pruritus that is so common and can develop in patients with other medical problems for reasons that are unrelated to their underlying conditions.
In my experience with patients like this, with multiple medical problems and therefore taking multiple medications, one of the most common causes of pruritus even without a rash is a drug allergy. The history of asthma and sinusitis suggests that he may be allergy-prone (atopic). Since he was on multiple medications, drug-induced pruritus was a reasonable cause. However, he did not start taking any new medication before the pruritus began, making it a little less likely that a drug allergy was causing the pruritus. Other than that, at this point, no etiology can be definitely excluded, and more information is needed.
Physical exam confirmed the history of multiple complaints and was significant for diffuse skin excoriations but no rash, prolonged expirations, unsteady gait, and sensorimotor deficits in both lower extremities. The peak expiratory flow rate was significantly reduced (330 L/min, predicted = 477 L/min). Skin testing with a battery of 14 common inhalant allergens, including many types of mold spores was negative.
The physical examination confirms a few facts and makes some diagnoses less likely. There is no obvious rash, indicating that the pruritus probably doesn’t have a primary dermatological cause, such as chronic urticaria, atopic dermatitis or psoriasis. On the other hand, the presence of excoriations indicates that the pruritus is probably quite intense. There is also significant neurological disease and some pulmonary disease as manifested by the airway obstruction. The negative skin testing makes an allergic cause only a little less likely.
Previous pulmonary function testing (PFT), which demonstrated decreased FEV1 and FVC, normal FEV1/FVC ratio, and no significant bronchodilator response (Figure). A chest x-ray revealed old inflammatory disease and no active disease (Figure) and a chest CT without contrast demonstrated bilateral lung nodules with diffuse ground-glass opacities (Figure).
The irreversible airway obstruction suggests chronic obstructive pulmonary disease and makes asthma a little less likely. However, the absence of a significant cigarette smoking or occupational exposure history argues against this diagnosis. The most important finding is the abnormal chest CT. This is the first piece of evidence that there is significant pulmonary disease other than asthma and it makes me a little more suspicious that there is some underlying multisystem problem. Again, more data is required.
Laboratory testing including serum chemistries, liver function tests, antinuclear antibody, antineutrophilic cytoplasm antibodies, HIV antibodies, sedimentation rate, urine/serum protein electrophoresis, serum IgA, IgG, and IgM were all within normal limits.
However a complete blood count was remarkable for marked leukocytosis (18.4 x 103/mm3) with 59% eosinophils (absolute - ?). Total serum IgE was very high (9562 IU/mL, normal < 180 IU/mL).
This pattern of laboratory abnormalities transforms this rather puzzling case with a number of potential diagnoses to one that has only a very few possibilities. The very high levels of eosinophils and IgE make it almost certain that these abnormalities are either directly related or actually causative of the pruritus. Although there are numerous causes of eosinophilia, there are few conditions of eosinophilia associated with an asthma-like presentation and pulmonary infiltrates. Those conditions include chronic eosinophilic pneumonia, acute bronchopulmonary aspergillosis (ABPA), Churg-Strauss syndrome and parasitic infections, including tropical pulmonary eosinophilia (TPE).
Churg-Strauss syndrome is an eosinophilic vasculitis with peripheral eosinophilia, but serum IgE levels of this magnitude are unusual in this condition. Conversely, ABPA is characterized by high levels of serum IgE but eosinophilia of this magnitude is unusual. The negative skin test to Aspergillus fumigatus and other molds also makes this diagnosis unlikely. Pulmonary disease and elevated serum IgE are also less common in the idiopathic hypereosinophilic syndrome and eosinophilic leukemia,
The combination of eosinophilia and elevated serum IgE strongly suggest some type of systemic allergic stimulation, such is seen with a parasitic infestation.
Review of his medical records indicated that he had eosinophilia for the past 7 years. His eosinophil count began increasing 7 years prior to our evaluation and had reached a plateau 2 years before. Most unusually, his eosinophil count continued to increase after he was started on oral corticosteroids three years previously, for the presumed CIPD (Figure).
This is a very important clue. Eosinophils are exquisitely sensitive to corticosteroids and eosinophilia of almost every etiology, except infestation with a viable parasite or eosinophilic leukemia, will decrease even on relatively low dose corticosteroid therapy.
Because of the eosinophilia and his Honduran background, he had been previously evaluated for parasitic infections. However, several stool analyses for ova and parasites were negative and a previous colonoscopy had revealed severe eosinophilic bowel inflammation, but no parasites.
Despite the previous evaluation, we felt that the patient was likely to have parasitic disease. The most common parasitic disease that we see in this setting is infection with Strongyloides stercoralis. Strongyloides infections are known to last for decades, so this patient could have been infected while residing in Honduras 35 years ago. Typically, these infections remain dormant until patients become immunocompromised. When this happens, strongyloidiasis hyperinfection syndrome develops, in which large numbers of recently released larvae burrow through the intestine and migrate to the lungs. The resultant large allergenic load stimulates the eosinophilia and increase in serum IgE. Commonly, the immunosuppression is iatrogenic, from corticosteroid therapy. Despite the active strongyloides infection, stool ova and parasite examinations are often negative and diagnosis is made through serological testing.
Parasite serologies demonstrated positive Strongyloides antibodies by enzyme-linked immunoassay (ELISA) (observed level = 77, positive > 8), positive antifilarial IgG1, and positive serum antifilarial antibody (observed level = 19.7 mg/mL, positive > 14.0 mg/mL) (Table ). Antifilarial IgG4, stool ova and parasites, and blood microfilariae levels were negative.
The patient was treated with ivermectin (200 mcg/kg/day) for 2 days as treatment for the presumed Strongyloides stercoralis infection. The pruritus rapidly and almost completely resolved and his respiratory symptoms improved significantly. Post-treatment PFT demonstrated improved FEV1 and FVC. One month after treatment, his eosinophil count had decreased to 5 percent and has remained at that level for 6 more months (Figure 1). Six months after treatment, the serum IgE had also dropped significantly to 1718, and by 10 months had dropped even further to 947 IU/ml (Figure). The only symptom that had not yet improved 10 months after ivermectin treatment is the polyneuropathy.
Commentary
This infection affects an estimated 30 million people worldwide, and is endemic to Southeast Asia, Latin America, sub-Saharan Africa, and the Southeastern United States but has a low prevalence in developed societies (1). It undergoes a complex life cycle within humans. Within the small intestinal mucosa of infected humans, adult S. stercoralis females lay eggs that hatch into rhabditiform larvae, which may exit the body via feces (1). Rhabditiform larvae may develop into either infective filariform larvae (which penetrate intact skin and initiate infection) or free-living adults (which mate and lay eggs from which rhabditiform larvae hatch). Infective filariform larvae travel through the circulation to the lungs. From the lungs, the filariform larvae enter the airway and are swallowed, and subsequently migrate to the small intestine where they become adults (1). The S. stercoralis life cycle is unique in that autoinfection may occur: rhabditiform larvae become infective filariform larvae, which penetrate either the skin of the perianal region or the intestinal mucosa; in both cases, the infective filariform larvae may reach the lungs and small intestine (as described above) or disseminate throughout the body (1). Autoinfection may result in chronic infections lasting several decades even in individuals who have not recently traveled to endemic areas. S. stercoralis infections have been reported to persist for as long as 65 years (2).
S. stercoralis infections may be asymptomatic or present with cutaneous, gastrointestinal, pulmonary, or systemic manifestations. S. stercoralis causes a motile, serpiginous, urticarial rash (i.e. larva currens) frequently seen on the buttocks, groin, and trunk (1). Gastrointestinal manifestations consist of diarrhea, indigestion, abdominal bloating, nausea, and anorexia (1). Pulmonary strongyloidiasis may present with cough, dyspnea, or a clinical state that resembles pneumonia or asthma (1). In hyperinfection syndrome, invasive filariform larvae cause leakage of gut flora from damaged bowel. This condition, which carries a mortality rate as high as 87%, may result in gram-negative septicemia, pneumonia, meningitis, and disseminated bacterial and fungal infections (1,3). Hyperinfection is most commonly seen in patients with immunosuppression, human T-lymphotropic virus (HTLV-1), kidney allograft recipients, and patients using corticosteroids (1). Due to the risk of this potentially fatal condition in immunosuppressed individuals, it is important to consider S. stercoralis in any patient who has traveled to endemic regions before administering steroid therapy. This patient illustrates the distinct organ systems (e.g. cutaneous, pulmonary, neurologic) that may be affected by S. stercoralis or associated eosinophilia (4).
The heterogeneous clinical presentations of S. stercoralis often lead to delays in diagnosis and treatment. S. stercoralis may be diagnosed by identifying larvae in stool samples. However, in most cases, the intestinal worm load is low and the excreted larvae are scant. Single stool examinations may only detect S. stercoralis in 30% of cases although repeated stool testing improves sensitivity. It has been reported that examining 7 serial stool samples leads to diagnostic sensitivity approaching 100% (5,6). Strongyloides serology is approximately 95% sensitive and 29% specific, while positive and negative predictive values are 30% and 95%, respectively (7). Positive results must be interpreted with caution as these serologies are known to cross-react with other parasitic infections (e.g. filariasis, schistosomiasis, and Ascaris lumbricoides) (8,9). Parasitic infections, including S. stercoralis, are also associated with increased serum levels of antigen-specific IgE (10). Recent data suggest that helminth-specific IgE may persist for years following treatment even in the absence of re-exposure to parasites (10). Chest radiographic findings associated with S. stercoralis are nonspecific, and may include pleural effusions or interstitial, alveolar, segmental, or lobar pulmonary infiltrates (11). This patient demonstrated the difficulties associated with microscopic identification of S. stercoralis. Although positive anti-filarial antibodies further complicated the clinical picture, these results may be explained by assay cross-reactivity. In the setting of a relevant medical history, chronic eosinophilia and positive serologies, treatment was initiated with favorable results.
The treatment of S. stercoralis is complicated by the organism’s unique ability to undergo autoinfection and chronically persist. The preferred therapy is ivermectin, which has been considered at least as effective as thiabendazole (1,12). Following prolonged treatment with either medication, an estimated 90 % of patients demonstrate resolution of presenting symptoms and eosinophilia. Ivermectin is recognized as having less frequent side effects than thiabendazole, which is associated with a relatively higher relapse rate and gastrointestinal symptoms (1,12). After treatment is administered, it is difficult to gauge when the organism has been completely eradicated due to the poor diagnostic sensitivity of stool testing. The optimal dosing schedule for ivermectin has yet to be determined (13).
Although disseminated strongyloidiasis is exceedingly rare in developed societies, clinicians must maintain a high index of suspicion in immunosuppressed patients and those being considered for corticosteroid therapy. This may become a more frequent problem as globalization increases travel between developed and undeveloped countries. Given the nonspecific clinical findings and difficulties in definitive diagnosis, S. stercoralis infectionpresents considerable diagnostic and therapeutic challenges. Strongyloidiasis may masquerade as other diseases, and can potentially result in septicemia and death. Early recognition of S. stercoralis infections based on clinical presentation, radiography, and laboratory studies should prompt immediate treatment initiation in order to avoid potentially catastrophic consequences.
Somewhat ironically, the patient felt so much better after the ivermectin treatment that he flew to Honduras for an extended vacation. Hopefully, he will not become re-infected with Strongyloides or any other parasite.