OBJECTIVE To identify clinical and serological differences of patients with reactive arthritis after infection with Lancefield group A β-haemolytic streptococci (GAS), compared with non-group A—that is, group C or G streptococci (NGAS:GCS/GGS), and a group of culture negative or unidentified streptococci (GUS).

RESULTS 41 patients (female/male ratio 22/19; mean (SD) age 38 (13) years) with reactive arthritis were included. Culture of throat swab was positive in 13 cases (32%): 6 (15%) GAS, 7 NGAS (17%), that is, 5 (12%) GCS, 2 (5%) GGS. In 28 cases throat culture remained negative resulting in a group of unidentified streptococci; antibiotic pre-treatment had been given by the general practitioner in 18 cases (64%). Arthritis was non-migratory, the number of arthritic joints in GAS and NGAS was similar, whereas in NGAS patients fewer joints were involved than in GUS: mean (SEM) 36 swollen joint index: 3.3 (1.0) in NGASv 5.6 (1.0) in GUS (p<0.005); 28 swollen joint index: 2.9 (1.0) in NGAS v 4.3 (0.8) in GUS (p<0.05). Extra-articular manifestations—that is, erythema nodosum/ multiforme, AV conduction block or hepatitis—were observed after GAS or GUS infection, but not after NGAS infection. ASO and/or antiDNase-B rose significantly in all patients. The maximal titres for ASO and antiDNase-B in 41 PSRA patients were: mean (SEM) 1242 (232) U/l and 890 (100) U/l respectively; the maximal ASO titres were similar in the three groups: mean (SEM) 1125 (185) in GAS, 625 (160) in NGAS (GAS v NGAS: p=0.17), and 1430 (320) U/l in GUS (NGAS v GUS: p=0.10). AntiDNase-B titres were: mean (SEM) 1075 (180) in GAS, 375 (105) in NGAS (GAS v NGAS: p<0.01), and 995 (125) U/l in GUS (NGAS v GUS: p<0.005). ASO: antiDNase-B ratios were: mean (SEM) 0.89 (0.21) in GAS, 2.60 (0.76) in NGAS (GAS v NGAS: p<0.05), and 1.43 (0.28) in GUS (NGAS v GUS: p=0.12).

CONCLUSION Post-streptococcal reactive arthritis occurs not infrequently. Differentiation of PSRA based on the causative streptococcal strain is frequently thwarted by negative throat cultures. Sometimes extra-articular manifestations are present that exclude NGAS as the causative organism. Serologically, lower antiDNase-B titres may be indicative for primary NGAS infection; the ASO/antiDNase-B ratio may be of additive value for differentiation in cases of a negative throat culture: the higher ASO/antiDNase-B ratios suggesting primary NGAS infection. In reactive arthritis, serological monitoring consisting of a simultaneous titration of antiDNase-B and ASO, seems to be of clinical importance to trace GAS induced cases, especially when throat cultures remain negative.