The pathophysiologic mechanisms of Guillain-Barré syndrome (GBS) associated with Zika virus (ZIKV) infection may be indicated by differences in clinical features. OBJECTIVE To identify specific clinical features of GBS associated with ZIKV infection. DESIGN, SETTING, AND PARTICIPANTS During the ZIKV epidemic in Puerto Rico, prospective and retrospective strategies were used to identify patients with GBS who had neurologic illness onset in 2016 and were hospitalized at all 57 nonspecialized hospitals and 2 rehabilitation centers in Puerto Rico. Guillain-Barré syndrome diagnosis was confirmed via medical record review using the Brighton Collaboration criteria. Specimens (serum, urine, cerebrospinal fluid, and saliva) from patients with GBS were tested for evidence of ZIKV infection by real-time reverse transcriptase-polymerase chain reaction; serum and cerebrospinal fluid were also tested by IgM enzyme-linked immunosorbent assay. In this analysis of public health surveillance data, a total of 123 confirmed GBS cases were identified, of which 107 had specimens submitted for testing; there were 71 patients with and 36 patients without evidence of ZIKV infection. Follow-up telephone interviews with patients were conducted 6 months after neurologic illness onset; 60 patients with and 27 patients without evidence of ZIKV infection participated. MAIN OUTCOMES AND MEASURES Acute and long-term clinical characteristics of GBS associated with ZIKV infection. RESULTS Of 123 patients with confirmed GBS, the median age was 54 years (age range, 4-88 years), and 68 patients (55.3%) were male. The following clinical features were more frequent among patients with GBS and evidence of ZIKV infection compared with patients with GBS without evidence of ZIKV infection: facial weakness (44 [62.0%] vs 10 [27.8%]; P < .001), dysphagia (38 [53.5%] vs 9 [25.0%]; P = .005), shortness of breath (33 [46.5%] vs 9 [25.0%]; P = .03), facial paresthesia (13 [18.3%] vs 1 [2.8%]; P = .03), elevated levels of protein in cerebrospinal fluid (49 [94.2%] vs 23 [71.9%]; P = .008), admission to the intensive care unit (47 [66.2%] vs 16 [44.4%]; P = .03), and required mechanical ventilation (22 [31.0%] vs 4 [11.1%]; P = .02). Six months after neurologic illness onset, patients with GBS and evidence of ZIKV infection more frequently reported having excessive or inadequate tearing (30 [53.6%] vs 6 [26.1%]; P = .03), difficulty drinking from a cup (10 [17.9%] vs 0; P = .03), and self-reported substantial pain (15 [27.3%] vs 1 [4.3%]; P = .03). CONCLUSIONS AND RELEVANCE In this study, GBS associated with ZIKV infection was found to have higher morbidity during the acute phase and more frequent cranial neuropathy during acute neuropathy and 6 months afterward. Results indicate GBS pathophysiologic mechanisms that may be more common after ZIKV infection.

No abstract

Guillain-Barré syndrome (GBS) is an uncommon autoimmune disorder characterized by progressive weakness and diminished deep tendon reflexes following infection or, rarely, vaccination. 1 Increased GBS incidence has been reported in countries affected by Zika virus, 2 a flavivirus transmitted primarily by Aedes species mosquitoes. 3 During the Zika virus epidemic in Puerto Rico, we conducted a casecontrol study to identify risk factors associated with GBS.Methods | This study was approved by the US Centers for Disease Control and Prevention and the University of Puerto Rico institutional review boards. Written informed consent was obtained from participants.Potential c ase-patients were reported to public health authorities from 9 reference hospitals throughout Puerto Rico with a clinical suspicion of GBS and neurologic illness onset from April 2016 through December 2016. Case-patients were offered enrollment within 1 month of reported neurologic illness onset. GBS diagnosis was retrospectively confirmed by chart review using the Brighton Collaboration criteria after hospital discharge. 4 Case-patients were matched to community controls 1:2 by age group (ie, 7-20, 21-39, 40-64, and >65 years) and place of residence (ie, ≤1-km radius from the residence of a casepatient). The control group included members of the community who lived continuously at the enrollment site for the previous 2 months and were enrolled within 1 week of the matched case-patient. Community control enrollment sites were identified using a randomly generated distance (ie, 0-1000 m) and degree from North (ie, 0°-359°) from each case-patient's residence.For all participants, a questionnaire was administered on demographics, behaviors, exposures, and medical history, including acute illness within the previous 2 months, and serum, urine, and saliva specimens were collected. Participants were defined as having acute Zika virus infection if they had a positive reverse transcription-polymerase chain reaction (RT-PCR) result in any specimen. Participants were defined as having laboratory evidence of Zika virus infection if they had a positive RT-PCR result in any specimen or anti-Zika virus immunoglobulin M (IgM) detected in serum by enzyme-linked immunosorbent assay.Using SAS (SAS Institute), version 9.3, the Pearson χ 2 test and 2 sample (2-sided) tests were used to analyze demographic variables. P values less than .05 were considered significant. Matched odds ratios (MORs) with 95% CIs were calculated to measure the association with risk factors, with CIs not including 1 considered significant.