Dear Sir:
Thank you for your interest and comments.1 In an earlier draft of our manuscript, we used less stringent variable inclusion criteria in the multivariable model (using a P value < 0.25). During the review and editing process, we decided that the large number of covariates relative to sample size resulted in an unstable outcome, with decreased generalizability beyond the study sample.2 Thus, we included only variables with P < 0.01. We note the limitation of this approach that you highlighted in your letter.
The multivariate analysis data in logistic regression (using a cutoff of P < 0.25) following univariate analysis are given in Tables 1 and 2.
Logistic regression analysis of factors associated with overall nephropathy in HIV-negative patients
| Variable | Normal to mildly increased albuminuria < 30 mg albumin/g creatinine | Nephropathy ≥ 30 mg albumin/g creatinine | Univariate analysis | Multivariate analysis | ||
|---|---|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | |||
| Sex | ||||||
|  Female | 132 | 97 | 1 | 0.744 | N/A | N/A |
|  Male | 45 | 36 | 1.09 (0.65–1.81) | |||
| Age, years, mean (SD) | 56.9 (15.1) | 59.8 (14.8) | 1.01 (1.00–1.03) | 0.089 | 1.01 (0.99–1.03) | 0.236 |
| Body mass index, kg/m2, mean (SD) | 27.4 (5.5) | 25.9 (4.9) | 0.95 (0.91–0.99) | 0.018 | 0.97 (0.92–1.02) | 0.285 |
| Duration of disease, years, mean (SD) | 9.0 (9.6) | 13.2 (11.5) | 1.04 (1.01–1.06) | 0.001 | 1.02 (0.99–1.05) | 0.129 |
| Consume alcohol | ||||||
|  Yes | 10 | 3 | 0.39 (0.10-1.43) | 0.154 | 0.20 (0.04-1.09) | 0.062 |
|  No | 167 | 130 | 1 | 1 | ||
| Taking alternative medicine | ||||||
|  Yes | 13 | 11 | 1.14 (0.49–2.63) | 0.763 | N/A | N/A |
|  No | 164 | 122 | 1 | |||
| HbA1c, %, median (IQR) | 7.8 (6.4–9.7) | 8.8 (7.4–11.0) | 1.21 (1.10–1.33) | < 0.001 | 1.13 (0.96–1.32) | 0.149 |
| Fructosamine, mmol/L | 3.3 (1.0) | 3.8 (1.3) | 1.00 (1.00–1.01) | < 0.001 | 1.00 (1.00–1.01) | 0.165 |
| Triglycerides, mmol/L | 1.2 (0.8–1.6) | 1.2 (0.8–1.2) | 1.11 (0.83–1.47) | 0.487 | N/A | N/A |
| Total cholesterol, mmol/L | 4.6 (3.7–5.4) | 4.5 (3.6–5.5) | 1.06 (0.91–1.23) | 0.456 | N/A | N/A |
| HDL cholesterol, mmol/L | 1.1 (0.9–1.4) | 1.1 (0.9–1.4) | 1.67 (0.92–3.01) | 0.091 | 1.25 (0.63–2.49) | 0.518 |
| Hypertension | ||||||
|  Yes | 152 | 121 | 1.66 (0.80–3.44) | 0.174 | 1.17 (0.45–3.01) | 0.752 |
|  No | 25 | 12 | 1 | 1 | ||
| Retinopathy | ||||||
|  Yes | 31 | 55 | 3.47 (2.06–5.87) | < 0.001 | 3.04 (1.70–5.45) | < 0.001 |
|  No | 141 | 72 | 1 | 1 | ||
CI = confidence interval; HDL = high-density lipoprotein; HIV = human immunodeficiency virus; IQR = interquartile range; N/A = not applicable; OR = odds ratio; SD = standard deviation. In univariate analysis, nephropathy in HIV-negative diabetic patients was significantly associated with lower body mass index (OR 0.95; 95% CI [0.91–0.99]), longer duration of disease (OR 1.04; 95% CI [1.01–1.06]), higher glycosylated hemoglobin (OR 1.21; 95% CI [1.10–1.33]), higher fructosamine (OR 1.00; 95% CI [1.00–1.01]), and retinopathy (OR 3.47; 95% CI [2.06–5.87]). When the variables were subjected to multivariate analysis, only retinopathy (OR 3.04; 95% CI [1.70–5.45]) remained a significant predictor of nephropathy.
Logistic regression analysis of factors associated with overall nephropathy in HIV-positive patients
| Variable | Normal to mildly increased albuminuria < 30 mg albumin/g creatinine | Nephropathy ≥ 30 mg albumin/g creatinine | Univariate analysis | Multivariate analysis | ||
|---|---|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | |||
| Sex | ||||||
|  Female | 9 | 12 | 1 | 0.483 | N/A | N/A |
|  Male | 4 | 9 | 1.69 (0.39–7.27) | |||
| Age, years, mean (SD) | 52.3 (13.4) | 53.1 (10.9) | 1.01 (0.95–1.07) | 0.835 | N/A | N/A |
| Body mass index, kg/m2, mean (SD) | 25.4 (5.7) | 24.8 (5.0) | 0.97 (0.85–1.11) | 0.700 | N/A | N/A |
| Duration of disease, years, mean (SD) | 7.2 (8.7) | 5.5 (6.5) | 0.97 (0.88–1.06) | 0.492 | N/A | N/A |
| Consume alcohol | ||||||
|  Yes | 0 | 1 | N/A | N/A | N/A | N/A |
|  No | 13 | 20 | ||||
| Taking alternative medicine | ||||||
|  Yes | 2 | 0 | N/A | N/A | N/A | N/A |
|  No | 11 | 21 | ||||
| HbA1c, %, median (IQR) | 7.1 (6.0–9.7) | 8.5 (6.0–10.4) | 1.16 (0.87–1.56) | 0.305 | N/A | N/A |
| Fructosamine, mmol/L | 2.7 (0.6) | 3.8 (1.6) | 1.01 (1.00–1.02) | 0.049 | 1.01 (1.00–1.02) | 0.048 |
| Triglycerides, mmol/L | 1.1 (0.7–1.3) | 1.4 (0.9–2.3) | 2.43 (0.76–7.81) | 0.136 | 3.07 (0.76–12.43) | 0.116 |
| Total cholesterol, mmol/L | 3.8 (3.0–4.7) | 4.5 (3.7–5.4) | 1.69 (0.90–3.18) | 0.101 | 1.43 (0.69–2.99) | 0.338 |
| HDL cholesterol, mmol/L | 1.1 (0.8–1.3) | 1.2 (0.8–1.5) | 1.35 (0.26–7.13) | 0.722 | N/A | N/A |
| Hypertension | ||||||
|  Yes | 11 | 17 | 0.77 (0.12–4.96) | 0.786 | N/A | N/A |
|  No | 2 | 4 | 1 | |||
| Retinopathy | ||||||
|  Yes | 2 | 5 | 1.67 (0.27–10.33) | 0.583 | N/A | N/A |
|  No | 10 | 15 | 1 | |||
CI = confidence interval; HDL = high-density lipoprotein; IQR = interquartile range; N/A = not applicable; OR = odds ratio; SD = standard deviation. In univariate analysis nephropathy in HIV-positive diabetic patients was significantly associated with higher fructosamine (OR 1.01; 95% CI [1.00–1.02]). When the variables were subjected to multivariate analysis, only higher fructosamine (OR 1.01; 95% CI [1.00–1.02]) remained a significant predictor of nephropathy. There was an increased odds of nephropathy with increase in fructosamine OR 1.004 (95% CI 1.001–1.007, P = 0.009), controlling for other variables in the model in the manuscript.
REFERENCES
- 1.↑
Pakzad R, Safiri S, 2017. Prevalence of and factors associated with nephropathy in diabetic patients attending an outpatients clinic in Harare, Zimbabwe: methodological issues. Am J Trop Med Hyg 97: 980.