Heart rate variability and length of stay in abdominal aortic surgery patients

Phyllis K. Stein*, Robert E. Schmieg, Ahmed El-Fouly, Peter P. Domitrovich, Timothy G. Buchman

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


Heart rate variability (HRV) can quantify autonomic nervous system function and may reflect clinical status. The predictive value of HRV in acute post-surgical patients is unknown. Methods: 24-hour Holter recordings were done in 92 abdominal aortic surgery patients (62M,30F,67±1 yrs) immediately post-op. Hourly power spectral plots were classified as virtually no HRV, normal-appearing HRV with respiratory sinus arrhythmia (RSA), and increased HRV with highly erratic heart rate (HR) and non-RSA (ESR). 24-hour time and frequency domain HRV and α (1/f slope of log frequency vs log power between 10-4 and 10-2 Hz, a non-linear power-law HRV index) were computed. Patients with length of stay (LOS) ≤ 8 days and > 8 days (median split) were compared. As ESR can elevate HRV, the group of patients with ≤2 hours of ESR (NORMHRV, N=57) was also examined. Data are expressed as mean±SEM. Results: Longer LOS was associated with greater age (70±2 vs 65±1 yrs, p=.02), more negative α (-1.41±.05 vs -1.18±.04)* and greater number of hours with no HRV (12±2 vs 5±1)*. In the NORMHRV group, these associations remained significant, and decreased short-term time and frequency domain HRV indices and higher HR were significantly associated with greater LOS. Non-NORMHRV patients were older (72±1 vs 65±2 yrs)* and had significantly higher short-term time and frequency domain HRV indices than the NORMHRV group. *(p<0.01, t-test). Conclusions: In abdominal aortic surgery patients, decreases in the non-linear power-law HRV index α in the immediate postop period are significantly associated with longer hospital stays. Standard time and frequency domain HRV measures were associated with length of stay only in patients with ≤2 hours of ESR. Erratic sinus rhythms, which artificially elevate HRV but do not reflect better autonomic function, confound the predictive value of HRV in postoperative patients. All Paltents (N=92) NORMHSV Patients (N=57) LOS ≤ 8 days LOS > 8day LOS ≤ 8 days LOS > 8 days N= 54 38 33 24 Age (years) 65±1 * 70±2 * 62±2 a 69±3 a Alpha -1.18±0.04 a -1.41±0.05 a -1.19±0.04 * -1.49±0.06 * no HRV (hrs) 5±1 * 12±2 * 6±2 * 15±2 * Data expressed as mean±SEM. T-test, a p<0.05, * p<0.01.

Original languageEnglish
Pages (from-to)A48
JournalCritical Care Medicine
Issue number12 SUPPL.
StatePublished - 1999
Externally publishedYes


Dive into the research topics of 'Heart rate variability and length of stay in abdominal aortic surgery patients'. Together they form a unique fingerprint.

Cite this