Patient characteristics for the combined series of 9 patients are summarized in Table 1 and Table 2, and detailed case descriptions for the 6 new patients are provided in the Technical Appendix. The patients included in this series were predominantly male (88.9%), and all but 1 were adults. Patients consumed raw crayfish while on float (recreational river) trips (7/9, 77.8%), camping (1/9, 11.1%), or as a demonstration of wilderness survival skills (1/9, 11.1%). Alcohol consumption at the time of crayfish consumption was common (7/9, 77.8%). Although there were differences in timing of seeking care and signs and symptoms, patients in this series frequently had cough (100%), fever (88.9%), and eosinophilia (100%).

Paragonimiasis can be difficult to diagnose in its early stages because of the nonspecific nature of initial symptoms. In some regions, paragonimiasis may be mistakenly diagnosed as tuberculosis. In this series of patients, initial diagnoses included pneumonia, bronchitis, influenza, gastroenteritis, acute cholecystitis, and pulmonary embolism. The median time between crayfish ingestion and the onset of clinical signs and symptoms was 4 weeks (range 2–12 weeks). The median interval between the onset of symptoms and the initial visit to health care facilities was 2 weeks (range 2–8 weeks). However, the median time from symptom onset to the correct diagnosis was 12 weeks (range 3–83 weeks). Before diagnosis of paragonimiasis, patients received multiple unnecessary medications and treatments, and these were sometimes associated with serious illness. All patients were treated with antimicrobial drugs.

Clostridium difficile infection developed in 1 patient after multiple courses of antimicrobial drug therapy. Six (67%) patients were treated with ≥l course of corticosteroids. One patient also underwent multiple thoracentesis procedures, and 1 of these procedures resulted in pneumothorax that required chest tube replacement. This patient also underwent decortication because of recurrent pleural effusions. One patient underwent laparoscopic cholecystectomy after having right upper quadrant pain. This finding may have been related to parasite migration across the diaphragm because the gallbladder did not show any pathologic changes.

Patients with paragonimiasis often have abnormal laboratory test results that are useful for making a diagnosis. Eosinophilia has been reported in 62%–66% of patients with infection caused by P. westermani flukes (20,21) and in 75% of patients with paragonimiasis in North America (19). All patients in this series had eosinophilia at initial examination (absolute eosinophil count range 600 cells/mm³–2,300 cells/mm³, % range 5.6%–21%). Pleural fluid analysis showed eosinophilia in 3 patients. Chest radiographic findings were abnormal for all patients with paragonimiasis in North America (19). Pleural effusions were present in 37% of paragonimiasis patients in Asia and in 60% of previously described patients in North America (19,22). All patients in this series had pleural effusions. Other chest radiographic findings included nodules, opacities, and infiltrates. Chest computed tomography scans showed pleural thickening, pericardial thickening, pericardial effusions, and worm nodules (23).

Four of 6 patients in the current series had pericardial effusions documented by either computed tomography or echocardiography. Although most pericardial effusions were small and did not cause hemodynamic compromise, 1 patient had cardiac tamponade that required emergency pericardiocentesis and drain placement. Analysis of pericardial fluid showed marked eosinophilia. Eosinophilic pericardial effusions were documented in 3 children with paragonimiasis caused by P. mexicanus flukes in Costa Rica (24,25). Pericardial effusion has also been reported for 1 patient with paragonimiasis in Asia (26). Pericardial effusions have not been reported for patients with P. kellicotti flukes infection, although various Paragonimus spp. flukes have been reported to invade soft tissue (19,20,27,28) and the central nervous system (19,29).

Serologic analysis can be useful for confirming a diagnosis of paragonimiasis. However, available serologic tests have limitations. An immunoblot for P. westermani flukes performed at the Centers for Disease Control and Prevention (Atlanta, GA, USA) has been reported to be highly sensitive (96%) and specific (99%) (30). However, this assay has not been validated for P. kellicotti flukes. In our series, 2 patients had negative immunoblot results at the Centers for Disease Control and Prevention for samples that had been positive by Western blot with P. kellicotti fluke antigen at Washington University (G.J. Weil, et al., unpub. data). These patients had symptoms and abnormal laboratory test results suggestive of paragonimiasis after ingestion of raw crayfish, and their symptoms resolved after therapy with praziquantel. Diagnosis by identification of ova in sputum specimens is specific, but has low sensitivity (30%–40%) (1). Ova were present in sputum from only 1 patient in our series (5). Examination of stool for ova has low sensitivity (11%–15%) (31,32).

Praziquantel (75 mg/kg in 3 divided doses for 2 days) is the treatment of choice for paragonimiasis in the United States (33). Cure rates of 71%–75%, 86%–100%, and 100% have been reported with 1-, 2-, and 3-day courses, respectively (1,34). All patients in this series were treated with praziquantel for 2–3 days, and 7 (77.8%) experienced rapid clinical improvement or cure after treatment. One patient had some residual dyspnea and chest tightness 4 weeks after treatment. These findings may have been related to the protracted time between onset of his symptoms and initiation of appropriate therapy. He was asymptomatic at the 6-month follow-up visit. One atypical patient with chronic paragonimiasis who also had preexisting chronic obstructive pulmonary disease did not notice much improvement in his chronic dyspnea after praziquantel treatment, but defervescence and a weight gain of 30 pounds represented a clear clinical response to therapy.

Control of this organism in the wild is not feasible because of the wide geographic distribution of crayfish and mammalian intermediate hosts that eat crayfish and serve as definitive hosts for the parasite. P. kellicotti flukes are highly prevalent among crayfish in rivers that are used for recreation in Missouri (5). Effective prevention strategies should focus on physician education to improve awareness of this disease and education targeted at the general population. We worked with public health officials to help improve awareness of this disease in physicians and in the general public. For example, we assisted the Missouri Department of Health and Senior Services in creating a health advisory (www.health.mo.gov/emergencies/ert/alertsadvisories/pdf/HAd4-30-10.pdf) for physicians in Missouri with the goal of educating physicians on the risk factors, clinical signs and symptoms, and treatment for this infection. In September 2009, we collaborated with the Missouri Department of Health and Senior Services and the Missouri Department of Natural Resources to create a warning poster (www.health.mo.gov/living/environment/fishadvisory/pdf/crayfish.pdf) that was posted at canoe rental facilities and campgrounds along rivers in Missouri. This poster warned the general public about the risk for consuming raw crayfish.

In addition, during the spring of 2010, four of the authors (M.A.L., L.M.D., T.C.B., G.J.W.) provided information to local and national print, radio, and television media to increase awareness of this infection. Three cases were identified after this media campaign. One patient sought care at our medical facility after his mother, a nurse, saw an article about paragonimiasis in her local newspaper. One patient was referred to our clinic by a friend who had seen a report on paragonimiasis on a local television station. Another patient had atypical features, but increased physician awareness helped to establish the diagnosis in this patient.