Latitudinal differences on the global epidemiology of infantile spasms: systematic review and meta-analysis

Jason L. Jia†,
Shiyi Chen†,
Vishalini Sivarajah,
Derek Stephens and
Miguel A. CortezEmail author View ORCID ID profile

†Contributed equally

Orphanet Journal of Rare Diseases201813:216

https://doi.org/10.1186/s13023-018-0952-x

Received: 3 September 2018
Accepted: 14 November 2018
Published: 29 November 2018

Abstract

Background

Infantile spasms represent the catastrophic, age-specific seizure type associated with acute and long-term neurological morbidity. However, due to rarity and heterogenous determination, there is persistent uncertainty of its pathophysiological and epidemiological characteristics. The purpose of the current study was to address a historically suspected latitudinal basis of infantile spasms incidence, and to interrogate a geographical basis of epidemiology, including the roles of latitude and other environmental factors, using meta-analytic and -regression methods.

Methods

A systematic search was performed in Ovid MEDLINE and Embase for primary reports on infantile spasms incidence and prevalence epidemiology.

Results

One thousand fifteen studies were screened to yield 54 eligible publications, from which 39 incidence figures and 18 prevalence figures were extracted. The pooled incidence was 0.249 cases/1000 live births. The pooled prevalence was 0.015 cases/1000 population. Univariate meta-regression determined a continental effect, with Europe demonstrating the highest onset compared from Asia (OR = 0.51, p = 0.004) and from North America (OR = 0.50, p = 0.004). Latitude was also positively correlated with incidence globally (OR = 1.02, p < 0.001). Sub-analyses determined a particularly elevated Scandinavian incidence compared to the rest of world (OR = 1.88, p < 0.001), and lack of latitudinal effect with Scandinavian exclusion (p = 0.10). Metrics of healthcare quality did not predict incidence. Multiple meta-regression determined that latitude was the key predictor of incidence (OR = 1.02, p = 0.001).

Conclusions

This is the first systematic epidemiological study of infantile spasms. Limitations included lack of Southern hemispheric representation, insufficient study selection and size to support some sub-continental analyses, and lack of accessible ethnic and healthcare quality data. Meta-analyses determined a novel, true geographical difference in incidence which is consistent with a latitudinal and/or ethnic contribution to epileptogenesis. These findings justify the establishment of a global registry of infantile spasms epidemiology to promote future systematic studies, clarify risk factors, and expand understanding of the pathophysiology.