Lack of evidence for interaction between tobramycin and shock in their effect on renal function

We sought to determine whether there was an interaction between aminoglycoside use and shock in their effect on renal function among seriously ill patients suspected of having gram-negative sepsis. Serial serum creatinine determinations were used to estimate changes in creatinine clearance rates in 179 patients entered onto a prospective randomized trial of tobramycin-nafcillin versus cefotaxime. A 25% decline in estimated creatinine clearance was considered to be clinically important. Univariate chi-square analysis showed that both shock (P less than 0.01) and tobramycin use (P less than 0.001) were independently associated with decline in estimated creatinine clearance. A two-way analysis of variance showed that both shock (F = 10.44, P less than 0.01) and tobramycin use (F = 42.6, P less than 0.001) continued to be significantly associated with renal dysfunction in the presence of each other, but there was no significant interaction between shock and tobramycin in their effect (F = 0.62, P less than 0.43). A multiple logistic regression with an interaction term representing the occurrence of shock and tobramycin use simultaneously yielded similar results. Our study provided no analytic evidence supporting the existence of an interaction between shock and aminoglycoside use in their effect on renal function.