Guillain-Barré Syndrome During Ongoing Zika Virus Transmission — Puerto Rico, January 1–July 31, 2016

Epidemiologic Surveillance for GBS

In February 2016, as part of the ongoing response to local Zika virus transmission (3–5), PRDH implemented GBPSS with CDC assistance. Health care providers throughout the island are requested to report all patients with suspected GBS to PRDH by submitting a case report form^† along with patient specimens (serum, urine, cerebrospinal fluid [CSF], and saliva). Reporting of patients with other neurologic disorders and a suspected antecedent arboviral infection is also encouraged. All submitted specimens are tested at CDC Dengue Branch (San Juan, Puerto Rico) or PRDH Biological and Chemical Emergencies Laboratory using a Trioplex reverse transcription–polymerase chain reaction (RT-PCR) assay to detect nucleic acid from Zika, dengue, and chikungunya viruses.^§ Serum and CSF specimens are also tested by immunoglobulin M (IgM) enzyme-linked immunosorbent assay (ELISA) for all three viruses. Persons with Zika virus nucleic acid detected by RT-PCR in any specimen are considered to have confirmed Zika virus infection. Persons with negative results for Zika virus by RT-PCR in all specimens tested are considered to have presumptive recent infection with a single arbovirus when results are positive for that virus by IgM ELISA, and presumptive flavivirus infection if they test positive for both Zika virus and dengue virus by IgM ELISA. Neurologic diagnosis for GBS cases with evidence of arboviral infection was confirmed by chart review using the Brighton Collaboration criteria, a set of standardized diagnostic criteria based on clinical presentation, CSF laboratory results, and electrophysiologic findings (6). Chart reviews are performed after hospital discharge, >28 days after onset of neurologic signs for persons who remain hospitalized, or death.

Fifty-six suspected cases of GBS with onset of neurologic signs during January 1–July 31, 2016, were identified, including one case identified before implementation of GBPSS and two cases identified during a mid-year assessment of the surveillance system. Among identified suspected cases of GBS, 20 patients (37%) had no evidence of Zika virus infection, two (4%) patients had pending laboratory test results, and 34 (61%) patients had any evidence of Zika virus or flavivirus infection. Among these 34 patients, 10 (18%) had confirmed Zika virus infection, 16 (29%) had presumptive Zika virus infection, and eight (14%) had presumptive flavivirus infection. Additionally, one (2%) patient had equivocal results for both Zika and dengue virus by IgM ELISA, and three (5%) had equivocal results for chikungunya virus by IgM ELISA.

During January–July 2016, the median monthly case count of persons with suspected GBS and no evidence of Zika virus infection or pending laboratory results, by month of onset of neurologic signs, was three (range = 1–5). Overall, the number of persons with suspected GBS and evidence of Zika virus or flavivirus infection was 2.5 times greater than the number of persons with suspected GBS and no evidence of Zika virus infection, with an increasing number of cases occurring each month beginning in April (Figure 1). The 34 patients with suspected GBS and evidence of Zika virus or flavivirus infection were residents of seven of eight PRDH health regions (Figure 2); the median age of these patients was 55 years (range = 21–88 years), and 20 (59%) of the 34 patients were female.

Charts have been reviewed for 32 of these 34 patients; GBS was confirmed for all 32. Acute illness during the preceding 2 months was recorded in the medical charts of 30 (94%) patients (Table). The most frequently reported symptoms associated with the reported antecedent acute illness were rash (n = 18, 53%), fever (n = 12, 35%), and diarrhea (n = 7, 21%). Median interval from antecedent acute illness to onset of neurologic signs was 5 days (range = 0–17 days). All 34 patients were hospitalized after onset of neurologic signs, and among patients who were discharged (n = 28) or had died (n = 1) as of August 18, median duration of hospitalization was 12 days (range = 6–47 days). Among the 32 patients with completed chart reviews, the most common recorded clinical neurologic signs and symptoms were hyporeflexia or areflexia (n = 31, 97%), leg weakness (n = 31, 97%), leg paresthesia (n = 24, 75%), arm weakness (n = 24, 75%), facial weakness (n = 20, 63%), arm numbness (n = 19, 59%), and dysphagia (n = 19, 59%). All 25 patients who underwent lumbar punctures had cytoalbuminologic dissociation (increased CSF protein concentration and total CSF white blood cell count <50 cells/μl), which is frequently observed in patients with GBS. Five patients had electrophysiologic testing, which indicated the acute inflammatory demyelinating polyneuropathy subtype of GBS in all five. All 34 patients were treated with IVIg. As of August 18, 21 (62%) of 34 patients had been admitted to intensive care units, and 12 (35%) required mechanical ventilation. As of August 18, five (15%) patients remained hospitalized, 15 (44%) were discharged home, and 13 (38%) were transferred to a rehabilitation or skilled nursing facility. One patient (3%) with presumptive flavivirus infection died of septic shock, despite treatment with broad-spectrum antibiotics.

Additionally, 26 cases of neurologic disorders other than GBS were reported to PRDH via GBPSS; seven (27%) patients had evidence of recent Zika virus or flavivirus infection. The neurologic disorders among these seven patients include three cases of encephalitis (one patient had confirmed Zika virus infection, one had presumptive Zika virus infection, and one had presumptive flavivirus infection); two cases of myelitis, both in patients with confirmed Zika virus infection; one case of acute neurologic deficit, in a patient with confirmed Zika virus infection; and one case of postinfectious papilledema, in a patient with presumptive Zika virus infection.

Surveillance System Assessment

During July 2016, the surveillance system’s completeness in ascertaining suspected cases of GBS was assessed. Thirty-two (46%) of 70 total hospitals on the island were contacted. Hospital staff members from medical records or infection control departments were asked to provide a list of patients who had been hospitalized during January–June 2016, and who had an International Classification of Disease, 10th revision code for GBS (G61.0) in their medical record. Unreported cases of possible GBS were defined as patients who were hospitalized after GBPSS was initiated in mid-February, had hospital stays >3 days, and were not reported to GBPSS. Patients with an alternative final diagnosis were excluded. Twenty-six (81%) of the 32 contacted hospitals responded, including 13 of the 16 hospitals that reported at least one suspected case of GBS and 13 hospitals from a purposive sample of 16 hospitals selected randomly among the remaining hospitals on the island after stratifying by hospital size and region.

Two cases of possible GBS that had not been reported directly to GBPSS were identified. Both had specimens submitted to PRDH’s Passive Arboviral Diseases Surveillance System shortly before onset of their neurologic illness, which confirmed Zika virus infection.^¶ The patients were hospitalized for GBS in May and June respectively, and GBS diagnosis was confirmed by chart review.

Discussion

Because of the increased incidence of GBS reported in countries affected by Zika virus (7), PRDH implemented GBPSS soon after reporting the first case of Zika virus disease (4). Timely reporting has allowed for identification of ten cases of GBS with Zika virus infection confirmed by RT-PCR. Before this report, only individual cases of confirmed Zika virus infection in patients with GBS have been described (8). Cases of GBS with evidence of flavivirus infection are expected to be attributable to Zika virus, the predominant flavivirus currently circulating in Puerto Rico; during November 1, 2015–July 7, 2016, PRDH identified 5,582 confirmed and presumptive Zika virus cases compared with only 136 cases of dengue virus infections (5). Although fatalities among GBS patients are rare (1), the potential severity and burden of GBS on the health care system are highlighted by the death described here and by the large proportion of patients that required intensive care services, including mechanical ventilation.

Consistent with the global epidemiology of GBS, the median age of patients with suspected GBS and evidence of Zika virus or flavivirus infection was >50 years (9). Although GBS generally occurs somewhat more frequently in men than in women (9), the majority of patients described in this report were females. Analyses to explore the unexpected distribution of cases by sex are planned, including an apparent predominance of women in Puerto Rico with Zika virus infections. Although provider outreach activities and GBS case clustering might partially explain increased monthly case counts during April–July, an increased number of cases is consistent with ongoing Zika virus transmission. In French Polynesia, where a Zika virus disease outbreak occurred during 2013–2014, an increase in the number of GBS patients was reported (7). In contrast to French Polynesia, where electrophysiologic studies (when performed) identified the acute motor axonal neuropathy subtype of GBS (7), in Puerto Rico, the acute inflammatory demyelinating polyneuropathy subtype of GBS was identified in all five GBS cases with evidence of Zika virus or flavivirus infection who had electrophysiologic studies performed. Consistent with other reports that have found a range of neurologic conditions potentially associated with Zika virus infection (10), 27% of the 26 patients with neurologic conditions other than GBS identified through the GBPSS had evidence of Zika virus or flavivirus infection.

Residents of and travelers to Puerto Rico and countries with ongoing Zika virus transmission should follow recommendations for prevention of Zika virus infections. Given the potential increase in GBS incidence during ongoing Zika virus transmission, health care providers in areas with ongoing local transmission should be familiar with the clinical features of GBS to ensure timely patient diagnosis and treatment. Patients with signs and symptoms consistent with GBS should seek medical attention, regardless of antecedent illness. Health care providers should report suspected cases of GBS and other neurologic conditions to public health authorities.**