Introduction: A systematic review and a meta-analysis of sleep studies in schizoaffective disorder were conducted using published articles researched in major databases within the period since inception to June 15, 2025. Sleep parameters of drug-free schizoaffective patients (SZA) were analyzed and, where available, compared with case-control data of healthy controls (HC), drug-free schizophrenics patients (SCZ) and drug-free depressed patients (MDD). Materials and methods: A total of 9 studies with 67 SZA, 88 SCZ, 79 HC and 131 MDD were included in the meta-analysis. Further 23 studies with 87 SZA, 143 SCZ, 1038 HC and 31 MDD were included in the systematic review. Data were extracted regarding 15 sleep parameters: total sleep time (TST), sleep latency (LAT), wakefulness time during sleep (WAKE), REM time (REMT), REM percentage (%REM), REM latency (REMLT), REM density (REMD), stage 1, 2 and 4 sleep time (ST1, ST2, ST4) and percentage (%ST1, %ST2, %ST4), delta sleep time (DELTAT) and percentage (ÞLTA). All the patients were drug-free for a minimum of two weeks in all but one study, where they have been drug-free for two days. A minimum of three studies was required to perform a meta-analysis for each parameter. The primary outcome was the standard mean difference. Data were fitted with a random-effects model. Publication bias assessment was checked by means of Egger's Regression and funnel plot asymmetry. Results: SZA showed reduced TST (p<0.001), increased LAT (p<0.001) and WAKE (p<0.001) compared to HC. SZA showed reduced REMT (p=0.006) and shortened REM latency (p=0.007) compared to HC. SZA showed reduced stage ST4 (p<0.001) and %ST4 (p<0.001) compared to HC. SZA did not differ from HC regarding %REM and DELTAT. SZA showed increased LAT compared to MDD (p<0.001). SZA did not differ from MDD regarding TST, WAKE, REMT, %REM, REML, REMD, ST4, %ST4, DELTAT, ÞLTA. SZA did not differ from SCZ regarding any sleep parameter. Conclusions: Drug-free SZA sleep does not seem to differ from that of drug-free SCZ and drug-free MDD (except for SLAT in this last comparison). SZA patients show a shorter and more fragmented sleep, with difficulties falling asleep, compared to HC. SZA show a shortened REML and a reduced REMT compared to HC, with slow-wave sleep time deficits like that of SCZ and MDD. Literature regarding sleep in SZA is very limited, therefore meta-analysis could not be performed for some parameters (REMD and ÞLTA in SZA vs HC; ST1, %ST1, ST2, %ST2 in SZA vs HC and in SZA vs MDD). Heterogeneity of 40% of meta-analyses in this study was substantially high (>70%). Further studies, preferably involving drug-naive patients, are needed to clarify whether SZA sleep is more similar to that of SCZ or MDD. Specific attention should be allocated to the investigation of REMD in SZA, for this data could help clarifying the significancy of REMD in human sleep and its usefulness as a marker of affective and/or psychotic disorders.
Sleep in schizoaffective disorder: a systematic review and meta-analysis
Morra, D.
Writing – Original Draft Preparation
;Ficca, G.Supervision
;Barbato, G.Supervision
2026
Abstract
Introduction: A systematic review and a meta-analysis of sleep studies in schizoaffective disorder were conducted using published articles researched in major databases within the period since inception to June 15, 2025. Sleep parameters of drug-free schizoaffective patients (SZA) were analyzed and, where available, compared with case-control data of healthy controls (HC), drug-free schizophrenics patients (SCZ) and drug-free depressed patients (MDD). Materials and methods: A total of 9 studies with 67 SZA, 88 SCZ, 79 HC and 131 MDD were included in the meta-analysis. Further 23 studies with 87 SZA, 143 SCZ, 1038 HC and 31 MDD were included in the systematic review. Data were extracted regarding 15 sleep parameters: total sleep time (TST), sleep latency (LAT), wakefulness time during sleep (WAKE), REM time (REMT), REM percentage (%REM), REM latency (REMLT), REM density (REMD), stage 1, 2 and 4 sleep time (ST1, ST2, ST4) and percentage (%ST1, %ST2, %ST4), delta sleep time (DELTAT) and percentage (ÞLTA). All the patients were drug-free for a minimum of two weeks in all but one study, where they have been drug-free for two days. A minimum of three studies was required to perform a meta-analysis for each parameter. The primary outcome was the standard mean difference. Data were fitted with a random-effects model. Publication bias assessment was checked by means of Egger's Regression and funnel plot asymmetry. Results: SZA showed reduced TST (p<0.001), increased LAT (p<0.001) and WAKE (p<0.001) compared to HC. SZA showed reduced REMT (p=0.006) and shortened REM latency (p=0.007) compared to HC. SZA showed reduced stage ST4 (p<0.001) and %ST4 (p<0.001) compared to HC. SZA did not differ from HC regarding %REM and DELTAT. SZA showed increased LAT compared to MDD (p<0.001). SZA did not differ from MDD regarding TST, WAKE, REMT, %REM, REML, REMD, ST4, %ST4, DELTAT, ÞLTA. SZA did not differ from SCZ regarding any sleep parameter. Conclusions: Drug-free SZA sleep does not seem to differ from that of drug-free SCZ and drug-free MDD (except for SLAT in this last comparison). SZA patients show a shorter and more fragmented sleep, with difficulties falling asleep, compared to HC. SZA show a shortened REML and a reduced REMT compared to HC, with slow-wave sleep time deficits like that of SCZ and MDD. Literature regarding sleep in SZA is very limited, therefore meta-analysis could not be performed for some parameters (REMD and ÞLTA in SZA vs HC; ST1, %ST1, ST2, %ST2 in SZA vs HC and in SZA vs MDD). Heterogeneity of 40% of meta-analyses in this study was substantially high (>70%). Further studies, preferably involving drug-naive patients, are needed to clarify whether SZA sleep is more similar to that of SCZ or MDD. Specific attention should be allocated to the investigation of REMD in SZA, for this data could help clarifying the significancy of REMD in human sleep and its usefulness as a marker of affective and/or psychotic disorders.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
