Perioperative and anesthesia-related cardiac arrests in geriatric patients: a systematic review using meta-regression analysis

Abstract

The worldwide population is aging, and the number of surgeries performed in geriatric patients is increasing. This systematic review evaluated anesthetic procedures to assess global data on perioperative and anesthesia-related cardiac arrest (CA) rates in geriatric surgical patients. Available data on perioperative and anesthesia-related CA rates over time and by the country's Human Development Index (HDI) were evaluated by meta-regression, and a pooled analysis of proportions was used to compare perioperative and anesthesia-related CA rates by HDI and time period. The meta-regression showed that perioperative CA rates did not change significantly over time or by HDI, whereas anesthesia-related CA rates decreased over time (P = 0.04) and in high-HDI (P = 0.015). Perioperative and anesthesia-related CA rates per 10,000 anesthetic procedures declined in high-HDI, from 38.6 before the 1990s to 7.7 from 1990-2017 (P < 0.001) and from 9.2 before the 1990s to 1.3 from 1990-2017 (P < 0.001), respectively. The perioperative CA rate from 1990-2017 was higher in low-HDI than in high-HDI countries (P < 0.001). Hence, a reduction in anesthesia-related CA rates over time was observed. Both perioperative and anesthesia-related CA rates only decreased with a high-HDI between time periods, and perioperative CA rates during 1990-2017 were 4-fold higher with low- compared to high-HDI in geriatric patients.

Figure 1

Flowchart of the process used to identify studies for inclusion.

Figure 2

Meta-regression of perioperative cardiac arrest rates according to time. Each circle represents a study and indicates its weight in the analysis. The correlation between perioperative cardiac arrests and time was not significant (slope: 0.0079; 95% CI: −0.0483 to 0.0642; P = 0.76).

Figure 3

Meta-regression analysis of perioperative cardiac arrest rates according to Human Development Index (HDI) status. Each circle represents a study and indicates its weight in the analysis. The correlation between perioperative cardiac arrests and HDI was not significant (slope: −1.0389; 95% CI: −6.7380 to 4.6601; P = 0.69).

Figure 4

Meta-regression of anesthesia-related cardiac arrest rates according to time. Each circle represents a study and indicates its weight in the analysis. (A) The correlation between anesthesia-related cardiac arrests and time was significant considering both low- and high-HDI countries (slope: −0.0699; 95% CI: −0.1394 to −0.0003; P = 0.04). (B) The correlation between anesthesia-related cardiac arrests and time was significant in only high-HDI countries (slope: −0.1049; 95% CI: −0.1762 to −0.0336; P = 0.015).

Figure 5

Meta-regression analysis of anesthesia-related cardiac arrest rates according to Human Development Index (HDI) status. Each circle represents a study and indicates its weight in the analysis. The correlation between anesthesia-related cardiac arrests and HDI was not significant (slope: 1.6188; 95% CI: −8.6021 to 11.8399; P = 0.71).