Background
In many low‐ to middle‐income countries (LMIC) assisted vaginal birth rates have fallen, while caesarean section (CS) rates have increased, with potentially deleterious consequences for maternal and perinatal mortality.

Aims
To review birth mode and perinatal mortality in a large LMIC hospital with strict labour management protocols and expertise in vacuum extraction.

Materials and Methods
We conducted a retrospective observational study at Port Moresby General Hospital in Papua New Guinea. Birth registers from 1977 to 2015 (39 years) were reviewed. Overall and modified (fresh stillbirths and early neonatal deaths ≥500 g) perinatal mortality rates (PMRs) were calculated by birthweight/birth mode.

Results
There were 365 056 births (5215 in 1977; 14 927 in 2015), of which 14 179 (3.9%) were vacuum extractions, 609 (0.2%) forceps births and 14 747 (4.4%) CS (increase from 2% to 5%). The failure rate of vacuum extraction was 2.5% (range 0.5–5.4%). Symphysiotomy was employed for 184 births.
From 1989 to 2015, the modified mean PMR for babies ≥2500 g was 8.1/1000 births (range 5.6–12.1; 6.9 in 2015), 9.1/1000 for babies ≥1500 g (7.3–14.8; 9.1 in 2015) and 7.5/1000 (0–21.7; 9.0 in 2015) for vacuum extractions (98% were ≥2500 g). The overall PMR for these years was 29.7/1000 births.

Conclusions
In an LMIC with rapidly increasing birth numbers a comparatively low PMR can be achieved while maintaining low CS rates. This may be in part accomplished through strict use of second‐stage protocols, perinatal audit, and supportive training that promotes judicious and proficient use of vacuum extraction and CS.