Volume 1, Number 4—October 1995
Dispatch
Does Treatment of Bloody Diarrhea due to Shigella dysenteriae Type 1 with Ampicilin Precipitate Hemolytic Uremic Syndrome?
Table 2
Dysentery patients | BD cases admitted to Samtah and Abut-Arish Hospitals | ||||||
---|---|---|---|---|---|---|---|
Antibiotic combination group | Antibiotic combinations | Total (N=110) | Developing | HUS rate (%) | Risk ration | 95% Confidence interval | |
Nalidixic acid with without other antibiotics but no ampicillan (b) | N(3/23), N+ G+N+C+A (0/1), M + N = C(0/2), N + C (0/4) | 41 | 3 | 28 | 7.3 | 1 | 0.23-4.34 |
No antibiotic | No antibiotic (0/12) | 12 | 0 | 6 | 0.0 | NC(e) | P value= 0.629(d) |
Antibiotic other than nalidixic acid or ampicillin | A(0/1),C(1/0), G(0/1), M+G(0/1) | 4 | 1 | 4 | 25.0 | 1.58 | 0.22-11.58(e) |
Ampicillan with or without other antibiotics but no nalidixic acid | 28 | 10 | 19 | 35.7 | 6.90 | 0.98-48.68 | |
Ampicillin only | P (3/3) | 6 | 3 | 5 | 50.0 | 6.11 | 1.31-28.54 |
Ampicillin with other antibiotics but no nalidixic acid | P+G(1/2),P+M (2/3),P+M+A 0/2), P+M+C (1/0) P+M+G (3/8) | 22 | 7 | 14 | 31.8 | 10.50 | 0.54-205.39 |
Ampicillin and nalidixic acid with or without other antibiotics | P+M+G+N(1/1), P+M+N(0/10), P+M+N+C(1/0). P+N(1/8), P+N+A(1/2) | 25 | 4 | 24 | 16.0 | 1.77 | 0.31-10.21 |
HUS rate(%) calculated form total number of BD in the five district hospitals using corresponding antibiotic.
a Mantel-Haensel weighted relative risk adjusted to hospital (Epi Info, version 6.02). Analysis restricted for data from Samtah and Abu-Arish hospitals. HUS was not reported for the other three district hospitals.
bReference group. A=amikacin, C=claforan, g=gentamicin, N=nalidixic acid, M=metronidazole, P=ampicillin. Numbers between parentheses in the second colunm (antibiotic combination indicate the (number of patients with HUS who took the corresponding antibiotic combination number or patients with BD who took the same antibiotic combination but did not develop HUS in the five district hospitals).
cNC=not calculated.
dOne-tailed Fisher's exact test.
eAdjusted Mantel-Haenszel relative risk could not be calculated.