Table 2.
Risk Ratios for the Development of Residual Calcifications After Antiparasitic Treatment in Patients With Viable Brain Parenchymal Neurocysticercosis
| Study Variable | Calcified/Cysts Resolved (%) | Univariate Models | Multivariate Model 1b | Multivariate Model 2c | |||
|---|---|---|---|---|---|---|---|
| RR (95% CI) | P Value | RR (95% CI) | P Value | RR (95% CI) | P Value | ||
| Cyst level | |||||||
| Cyst size, mm | |||||||
| ≤14 | 166/453 (36.6) | Ref. | Ref. | Ref. | |||
| >14 | 22/44 (50.0) | 1.35 (1.14–1.60) | <.001 | 1.30 (1.07–1.57) | .007 | 1.34 (1.02–1.75) | .035 |
| Cyst content | |||||||
| Clear | 158/435 (36.3) | Ref. | |||||
| Turbid | 30/62 (48.4) | 1.30 (.18–2.15) | .313 | ||||
| Cyst location | |||||||
| Frontal lobe | 67/172 (39.0) | 1.02 (.78–1.35) | .866 | ||||
| Parietal lobe | 60/182 (33.0) | .82 (.67–1.00) | .047 | ||||
| Occipital lobe | 27/63 (42.9) | 1.13 (.85–1.52) | .402 | ||||
| Temporal lobe | 36/98 (36.7) | .98 (.70–1.38) | .902 | ||||
| Cyst edema | |||||||
| No | 125/365 (34.3) | Ref. | Ref. | Ref. | |||
| Yes | 63/132 (47.7) | 1.37 (1.17–1.62) | <.001 | 1.36 (1.16–1.59) | <.001 | 1.39 (1.05–1.85) | .023 |
| Patient level | |||||||
| Age (tertiles), y | |||||||
| 16–25 | 67/177 (37.9) | Ref. | |||||
| 26–35 | 56/156 (35.9) | .98 (.69–1.38) | .899 | ||||
| >35 | 65/164 (39.6) | 1.03 (.81–1.32) | .796 | ||||
| Sex | |||||||
| Female | 63/172 (36.6) | Ref. | |||||
| Male | 125/325 (38.5) | 1.04 (.83–1.30) | .733 | ||||
| Disease time, mos | |||||||
| ≤24 | 79/229 (34.5) | Ref. | Ref. | ||||
| >24 | 109/268 (40.7) | 1.17 (1.13–1.21) | <.001 | 1.25 (1.08–1.46) | .003 | ||
| Number of cysts | |||||||
| 1–2 | 38/78 (48.7) | Ref. | Ref. | ||||
| 3 or more | 150/419 (35.8) | .74 (.44–1.25) | .261 | ||||
| Number of calcifications at baseline | |||||||
| None | 39/98 (39.8) | Ref. | |||||
| 1 or more | 149/399 (37.3) | .95 (.55–1.69) | .891 | ||||
| Electroimmunotransfer blot bands | |||||||
| ≥4 | 140/388 (36.1) | Ref. | Ref. | Ref. | |||
| ≤3 | 48/109 (44.0) | 1.21 (1.14–1.28) | <.001 | 1.14 (1.02–1.27) | .020 | ||
| Previous APT | |||||||
| No | 173/440 (39.3) | Ref. | |||||
| Yes | 15/57 (26.3) | .69 (.36–1.30) | .249 | ||||
| Previous antiepileptic drugs | |||||||
| No | 92/254 (36.2) | Ref. | |||||
| Yes | 96/243 (39.5) | 1.11 (.99–1.23) | .064 | ||||
| APT scheme | |||||||
| ABZ + Praziquantel | 78/224 (34.8) | Ref. | Ref. | ||||
| Standard ABZ | 73/190 (38.4) | 1.09 (.84–1.41) | .525 | 1.04 (.71–1.51) | .861 | ||
| Increased ABZ | 37/83 (44.6) | 1.27 (1.18–1.36) | <.001 | 1.26 (1.14–1.39) | <.001 | ||
| Days with APT | |||||||
| 10 | 136/361 (37.7) | Ref. | Ref. | ||||
| 14 | 52/136 (38.2) | .99 (.88–1.13) | .958 | ||||
| Dose of DXM, mg/d | |||||||
| >6.5 | 71/217 (32.7) | Ref. | Ref. | Ref. | Ref. | ||
| ≤6.5 | 117/280 (41.8) | 1.26 (1.04–1.52) | .018 | 1.36 (1.02–1.81) | .037 | ||
| Days with DXM | |||||||
| 10–12 | 163/429 (38.0) | Ref. | Ref. | ||||
| 28 | 25/68 (36.8) | .95 (.86–1.06) | .366 | ||||
| Cyst cure and retreatment | |||||||
| Incomplete cure and retreated | 19/66 (28.8) | Ref. | Ref. | Ref. | Ref. | ||
| Complete cure and not retreated | 111/280 (39.6) | 1.41 (1.21–1.63) | <.001 | 1.48 (1.29–1.71) | <.001 | ||
| Incomplete cure and not retreated | 58/151 (38.4) | 1.33 (1.16–1.53) | <.001 | 1.45 (1.08–1.93) | .012 |
The empty cells correspond to variables that were not not selected for multivarate models 1 and 2.
Abbreviations: ABZ, albendazole; APT, antiparasitic treatment; CI, confidence interval; DXM, dexamethasone; RR, risk ratio.
aRisk ratios from univariate and multivariate models were estimated using multilevel Poisson regression with random effects for individual intercepts.
bMultivariate multilevel Poisson regression model 1 initially included all covariates at the cyst level using a stepwise approach with backward selection algorithm to retain covariates with P value < .05 for the Wald Test.
cMultivariate multilevel Poisson regression model 2 included covariates retained in model 1 and all covariates at the patient level using a similar stepwise approach for selection of covariates.