From the Department of Surgery, Division of Plastic Surgery and Yale Child Study Center, Yale School of Medicine.
Department of Plastic Surgery, University of Pittsburgh Medical Center.
Plast Reconstr Surg. 2024 Oct 1;154(4):824-835. doi: 10.1097/PRS.0000000000010999. Epub 2023 Aug 15.
Radiographic severity of metopic synostosis has been suggested as a predictor of long-term neurocognitive outcomes, and artificial intelligence (AI) has recently been used to quantify severity. Age at surgery is predictive of long-term neurocognition in sagittal synostosis but has not been adequately explored in metopic synostosis.
Children ages 6 to 18 years with corrected metopic synostosis underwent testing of intelligence quotient, academic achievement, and visuomotor integration (VMI). Various manual measurements and AI-derived severity scores were determined. Scans were categorized as moderate or severe for head-to-head comparisons and multivariable linear regressions were used to assess the relationship of age at surgery and severity with neurocognitive outcomes.
A total of 41 patients with average age at testing of 10.8 ± 3.4 years were included. A total of 18 patients were in the severe group and 23 patients were in the moderate group, with average ages at surgery of 6.6 ± 2.7 and 10.6 ± 8.4 months, respectively ( P = 0.062). Greater AI-derived severity was significantly associated with lower reading comprehension ( P = 0.040 and 0.018) and reading composite scores ( P = 0.024 and P = 0.008). Older age at surgery was significantly associated with lower VMI scores ( P values ranging from 0.017 to 0.045) and reading composite scores ( P = 0.047 and 0.019).
This study suggests an association between greater AI-derived radiographic severity and lower reading ability in corrected metopic synostosis. Older age at surgery was independently associated with lower reading ability and VMI. Surgical correction may mitigate neurodevelopmental differences based on severity that have been observed preoperatively.
CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, II.
额骨矢状缝早闭的放射学严重程度被认为是长期神经认知结果的预测因素,最近人工智能(AI)已被用于量化严重程度。手术年龄是矢状缝早闭长期神经认知的预测因素,但在额骨矢状缝早闭中尚未得到充分探讨。
6 至 18 岁患有矫正性额骨矢状缝早闭的儿童接受智商、学业成绩和视动整合(VMI)测试。确定了各种手动测量和 AI 衍生的严重程度评分。对头对头比较进行了扫描分类为中度或重度,并使用多变量线性回归来评估手术年龄和严重程度与神经认知结果的关系。
共纳入 41 例平均年龄为 10.8 ± 3.4 岁的患者。共有 18 例患者为重度组,23 例患者为中度组,手术年龄分别为 6.6 ± 2.7 个月和 10.6 ± 8.4 个月(P = 0.062)。AI 衍生的严重程度越大,阅读理解(P = 0.040 和 0.018)和阅读综合成绩(P = 0.024 和 P = 0.008)越低。手术年龄越大,VMI 评分(P 值范围从 0.017 到 0.045)和阅读综合成绩越低(P = 0.047 和 0.019)。
本研究表明,在矫正性额骨矢状缝早闭中,AI 衍生的放射学严重程度越大,阅读能力越低。手术年龄越大,阅读能力和 VMI 越低。手术矫正可能会减轻术前观察到的基于严重程度的神经发育差异。
临床问题/证据水平:风险,II。