2026 ECNP Congress
PS04-4020NMDA antagonist models for antipsychotic testing in rats: a systematic review and meta-analysis
Background
A neurobiological hypothesis links schizophrenia symptoms to glutamatergic dysfunction. NMDA receptor (NMDAR) antagonists (ketamine, MK-801, PCP) are widely used to model schizophrenia-relevant behaviours in rodents. However, variability in dosing, behavioural assays, and study design may limit reproducibility and translational validity. This study systematically evaluated behavioural outcomes in NMDAR antagonist models in rats.
Methods
A systematic review and meta-analysis were conducted following PRISMA and SYRCLE guidelines (PROSPERO: CRD420251159102). PubMed, Embase, Web of Science, and PsycINFO (2006–2025) were searched for rat studies using ketamine, MK-801, or PCP. Studies were stratified into acute and sub-chronic paradigms. Standardised mean differences (SMDs) with 95% confidence intervals (95% CI) were calculated using random-effects models, with heterogeneity assessed using I².
Results
A total of 670 studies were identified (410 acute; 260 sub-chronic). Acute studies were predominantly male (95%) and mainly used MK-801 (52%), whereas sub-chronic studies included more females (22%) and mainly used PCP (43%). The meta-analysis results are summarised in Table 1. Acute paradigms showed large effects in positive-like behaviours, particularly locomotor activity (LMA) and prepulse inhibition (PPI), driven by MK-801 and PCP. Cognitive and social outcomes were more variable, although social interaction (SIT) and novel object recognition (NOR) deficits were observed across antagonists. Sub-chronic paradigms showed reduced positive-like effects, particularly non-significant effects of PCP, but consistent deficits in SIT and NOR, with ketamine showing the largest effects. NOR was the most consistent cognitive endpoint across studies.
Conclusions
NMDAR antagonist models produce robust, paradigm-dependent behavioural effects. Acute models preferentially capture positive-like phenotypes, whereas sub-chronic paradigms better model stable cognitive and social deficits. However, substantial heterogeneity limits reproducibility, underscoring the need for improved standardisation and model selection.
Table 1.
| Acute | Sub-chronic | |||||
|---|---|---|---|---|---|---|
| SMD (95% CI) | N | I2 | SMD (95% CI) | N | I2 | |
| LMA | ||||||
| Ketamine | 1.83 (1.45 , 2.22)*** | 30 | 71% | 1.68 (1.18 , 2.18)*** | 35 | 87% |
| MK-801 | 2.63 (2.27 , 3.00)*** | 71 | 79% | 0.82 (0.35 , 1.29)*** | 33 | 81% |
| PCP | 2.11 (1.77 , 2.44)*** | 45 | 68% | 0.21 (-0.01 , 0.43) | 20 | 37% |
| Overall | 2.27 (2.05 , 2.48)*** | 146 | 75% | 0.96 (0.70 , 1.22)*** | 88 | 83% |
| PPI | ||||||
| Ketamine | -1.50 (-2.00 , -1.01)*** | 15 | 66% | -2.75 (-3.88 , -1.63)*** | 9 | 83% |
| MK-801 | -2.03 (-2.31 , -1.75)*** | 44 | 59% | -1.50 (-2.20 , -0.80)*** | 13 | 78% |
| PCP | -2.63 (-3.16 , -2.11)*** | 17 | 65% | -0.91 (-1.67 , -0.15) | 8 | 79% |
| Overall | -2.06 (-2.30 , -1.83)*** | 76 | 66% | -1.66 (-2.15 , -0.15)** | 30 | 81% |
| SIT | ||||||
| Ketamine | -1.84 (-3.29 , -0.39)* | 8 | 91% | -1.66 (-2.25 , -1.07)*** | 26 | 85% |
| MK-801 | -2.82 (-3.57 , -2.07)*** | 12 | 76% | -1.49 (-2.03 , -0.95)*** | 16 | 73% |
| PCP | -2.49 (-3.72 , -1.27)*** | 4 | 71% | -1.66 (-2.62 , -0.71)*** | 16 | 90% |
| Overall | -2.46 (-3.12 , -1.79)*** | 24 | 85% | -1.59 (-1.98 , -1.20)*** | 58 | 85% |
| NOR | ||||||
| Ketamine | -2.52 (-3.24 , -1.80)*** | 18 | 82% | -2.33 (-3.08 , -1.59)*** | 18 | 81% |
| MK-801 | -1.80 (-2.24 , -1.36)*** | 24 | 75% | -1.46 (-1.92 , -0.99)*** | 15 | 58% |
| PCP | -1.73 (-2.77, -0.69)** | 8 | 86% | -1.49 (-1.72 , -1.26)*** | 49 | 56% |
| Overall | -2.02 (-2.38 , -1.66)*** | 50 | 81% | -1.60 (-1.82 , -1.39)*** | 82 | 66% |
N: number of studies.**p<0.05; **p<0.01; **p<0.001
References
No references