BACKGROUND: Most (>80%) invasive GBS (iGBS) disease occurs in non-pregnant adults. Some individuals have recurrent iGBS. We examined and compared clinical and epidemiological characteristics of recurrent vs. non-recurrent iGBS among U.S. adults.
METHODS: CDC’s Active Bacterial Core Surveillance, is a network of 10 U.S. sites; of these, 9 conduct population-based surveillance for iGBS among adults (isolated from a normally sterile site). Recurrent cases were defined as occurring >30 days and ≤2 years after initial GBS culture. Medical records were reviewed for clinical information. This analysis included cases reported from 1996-2011 who survived initial infection. The Cochran-Armitage test for trend was used to assess changes in proportions by 1-year intervals.
RESULTS: Analysis of 23,140 iGBS cases in non-pregnant adults who survived initial infection demonstrated that iGBS incidence increased from 4.8 to 8.6 cases/100,000 persons (x2 for trend: p < 0.001) during 1996-2011. Among these cases, 763 (3.3%) had ≥1 recurrence: 46% ≤6 months and 88% ≤18 months. Among recurrent cases, 89 had 3, 20 had 4-5, and 3 had >5 infections within 2 years. Recurrent GBS incidence increased from 0.17 to 0.28 cases/100,000 persons during 1996 to 2011 (x2for trend: p < 0.001). In 1997, 44% recurrent cases were <65 years; by 2011, 59% were aged <65 years at time of initial culture (trend: z= -2.5, 2-sided p = 0.01). Obesity at initial infection was reported in 33% of recurrent iGBS cases versus 24% with non-recurrent disease (x2: p = 0.003). Additionally, the proportion of recurrent iGBS cases with obesity increased over time (38% in 2005 to 53% in 2011 trend: z= 2.7, 2-sided p < 0.001), but not the proportion with diabetes, heart disease, or immunosuppression. There were no differences in these co-morbidities in recurrent vs. non-recurrent iGBS. The most common syndromes among recurrent and non-recurrent cases were primary bacteremia, cellulitis, and osteomyelitis/septic arthritis. Among recurrent cases, 38% (289/763) had a different presentation during their second infection. The proportion of recurrent cases with osteomyelitis/septic arthritis as initial infection increased over time; 14% (1996) to 24% (2011) (trend: z= 2.3, 2-sided p = 0.02).
CONCLUSIONS: Incidence of both primary and recurrent iGBS increased over 16 years. Among recurrent cases, age decreased, prevalence of obesity and the frequency of bone/joint infections at the time of initial GBS infection increased. Further investigation of risk factors for recurrence may help inform prevention measures.