The Effect of Anticholinergic Use on the Neurocognitive Scores of Pediatric Spina Bifida Patients
Soo Jeong Kim, MD1, Angeline Johny, MD2, Alan Hsieh, MD2, Jonathan Porter Castillo, MD1, Heidi Castillo, MD1, Vinaya Bhatia, MD1, Alex Borden, PA-C1, Paul Austin, MD1.
1Texas Children's Hospital, Houston, TX, USA, 2Baylor College of Medicine, Houston, TX, USA.
Word Count: 307 + 3 tables (457; limit 500)
Title: The Effect of Anticholinergic Use on the Neurocognitive Scores of Pediatric Spina Bifida Patients
Authors: Soo Jeong Kim, Angeline Johny, Alan Hsieh, Jonathan Porter Castillo, Heidi Castillo, Vinaya Bhatia, Alex Borden, Paul F. Austin
Presenting author: Soo Jeong Kim
Background: In elderly patients, there is an association between anticholinergics use and cognitive decline or dementia. This study aims to characterize the cognitive effects of anticholinergic use on children with spina bifida, with the hypothesis that anticholinergic use has a negative impact on neurocognitive scores.
Methods: We performed an IRB-approved retrospective review of pediatric patients (0 to 18 years of age at presentation) followed at a high-volume tertiary care center for spina bifida. The results of neurocognitive testing (Capute scales [CAT/CLAMS], Kaufman Brief Intelligence Test-2nd edition [KBIT-2], Beery-Buktenica Developmental Test of Visual-Motor Integration [VMI]) were compared between patients who received or did not receive anticholinergics. Total days of anticholinergic exposure (quantified using prescription information), lesion level, and shunted hydrocephalus status were included in multiple linear regression analysis.
Results: 396 patients met inclusion criteria. 264 (67%) of patients were prescribed anticholinergics for a range of 4-5,411 days (median 983 days) (Table 1). Patients who received anticholinergics were more likely to have hydrocephalus requiring shunt and higher lesion levels, although the latter was not significantly different. Higher lesion levels and hydrocephalus requiring shunt were significantly associated with worse CAT/CLAMS, KBIT-2, and VMI scores (Table 2 & 3). On multiple linear regression analysis, higher total dose of anticholinergics was associated with significantly worse neurocognitive scores for the CAT/CLAMS test (CAT: 0.013 decrease in score per additional day of anticholinergic use; p=0.005; CLAMS: 0.01 decrease in score per additional day of anticholinergic use; p=0.03) but not for KBIT-2 or VMI.
Conclusions: Exposure to anticholinergics was associated with lower scores on CAT/CLAMS testing. Limitations include the possibility that these neurocognitive tests did not capture the domains potentially affected by anticholinergic medications. Alternatively, anticholinergic usage could have compounded the negative effects of variables like higher lesion levels and hydrocephalus requiring shunt. In order to address these confounding variables, further prospective studies are necessary.
Table 1 Demographics
| No anticholinergic(N=132) | Anticholinergic(N=264) | p-value | |
| Female gender | 72 (55%) | 136 (52%) | 0.57 |
| Length of follow-up in years (mean [standard deviation]) | 4.4 [3.8] | 7.9 [4.6] | <0.001 |
| Shunted hydrocephalus | 58 (44%) | 172 (65%) | 0.05 |
| Shunt revision | 22/58 (38%) | 111/172 (65%) | |
| Lesion level | 0.11 | ||
| Cervical | 0 | 0 | |
| Thoracic | 14 (11%) | 33 (13%) | |
| Lumbar | 63 (48%) | 151 (57%) | |
| Sacral | 52 (39%) | 79 (30%) | |
| Unknown | 2 (2%) | 1 (1%) | |
| Neurocognitive test | |||
| CAT/CLAMS (median age: 34 mos) | 84 (64%) | 75 (28%) | <0.001 |
| VMI (median age: 105.8 mos) | 33 (25%) | 131 (50%) | <0.001 |
| KBIT-2 (median age: 96.6 mos) | 29 (22%) | 97 (37%) | 0.003 |
| Days of anticholinergic use (range; median) | |||
| CAT/CLAMS | - | 4-2,723 (672) | - |
| VMI | - | 21-3,284 (996) | - |
| KBIT-2 | - | 8-5,411 (1,137) | - |
Table 2 Hydrocephalus requiring shunts significantly affects neurocognitive scores
| Scores | No shunt | Hydrocephalus requiring shunt | P-values | ||
| N | Mean ± SD | N | Mean ± SD | ||
| CLAM | 83 | 81.7 ± 24.3 | 76 | 63.2 ± 29.4 | <.0001 |
| CAT | 83 | 79.6 ± 26.6 | 76 | 64 ± 28.2 | <.0001 |
| VMI | 49 | 85.2 ± 23.3 | 115 | 75.3 ± 19.3 | .0002 |
| KBIT Composite | 43 | 80.5 ± 36.4 | 82 | 70.1 ± 28.7 | .045 |
Table 3 Lesion level significantly affects neurocognitive scores
| Scores | Thoracic | Lumbar | Sacral | P-values | |||
| N | Mean ± SD | N | Mean ± SD | N | Mean ± SD | ||
| CLAM | 14 | 59.3 ± 34.3 | 87 | 69.7 ± 28.6 | 58 | 80.7 ± 24.6 | .0057 |
| CAT | 14 | 53.9 ± 27.9 | 87 | 72.5 ± 27.2 | 58 | 76 ± 29.1 | .016 |
| VMI | 21 | 64.1 ± 14.9 | 95 | 78.8 ± 21 | 48 | 83.2 ± 20.9 | .0001 |
| KBIT Composite | 13 | 53.6 ± 30.6 | 72 | 73.1 ± 30.8 | 40 | 81.4 ± 31.7 | .017 |
Back to 2021 Posters