番茄社区

Preoperative hyponatremia significantly increased the risk of 30-day postsurgical mortality, a risk that extended across all subgroups of surgical patients, investigators reported.

Four times as many patients with preoperative hyponatremia died within 30 days of surgery as compared with patients who had sodium levels in the normal range before surgery. An adjusted analysis resulted in an overall odds ratio of 1.44, even higher in patients undergoing elective surgery and ASA class 0 and 1 patients, as reported online in Archives of Internal Medicine.

Hyponatremia also increased the risk of perioperative coronary events, wound infection, and pneumonia, and added a day to the average length of stay, wrote David W. Bates, MD, of Brigham and Women's Hospital in Boston, and co-authors.

"Hyponatremia, when detected preoperatively, should be considered a prognostic marker for perioperative complications, and its presence should alert physicians to a situation of increased risk necessitating closer surveillance in the perioperative period," they wrote in conclusion.

"Although the effectiveness and safety of intervening on preoperative hyponatremia have not yet been established, one reasonable approach is to monitor for perioperative complications in all patients at risk and to selectively treat hyponatremia before non-emergency surgical procedures when a reversible cause is found," they added.

Hyponatremia is a common finding in hospitalized patients and has potential consequences for a variety of medical conditions, including increased mortality risk, prolonged length of stay, and increased resource utilization and cost.

Previous studies of hospitalized patients with hyponatremia have focused on patients admitted to medical services. The relationship between preoperative hyponatremia and perioperative outcomes has remained largely unexplored, providing a rationale for the study undertaken by Bates and colleagues.

Data for the study came from the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) participant files, which comprise records for about 300 academic and community hospitals in the U.S.

Investigators searched records for all surgical cases reported to NSQIP from Jan. 1, 2005 to Dec. 31, 2010. They excluded cases directly related to acute trauma, transplantation, procedures involving brain-death organ donors, and concurrent cases.

Hyponatremia was defined as sodium levels <135 mEq/L, and was further categorized as mild (130 to 134 mEq/L) or moderate/severe (<130 mEq/L). The primary outcome was 30-day mortality.

The final analysis encompassed 964,263 patients, including 75,423 (7.8%) with preoperative hyponatremia, which was mild in 89% of cases. Hyponatremia occurred most often in cardiac and vascular surgery patients (11.8%, 11.2%, respectively), followed by general surgery (7.5%), and orthopedic surgery (7.1%).

Hyponatremic patients tended to be older, male, sicker, inpatients, and individuals requiring emergency surgery. Sodium measurements occurred within 1 day of surgery in 95% of emergency cases and 30.6% of non-emergency cases.

Normal preoperative sodium levels were associated with a 30-day mortality of 1.3%, increasing to 4.6% for patients with mild hyponatremia (adjusted OR 1.38, P<0.001) and to 9.6% for the 8,546 patients with moderate/severe hyponatremia (aOR 1.72, P<0.001).

Preoperative hyponatremia also increased the risk of adverse events:

• Major coronary events -- 1.8% versus 0.7%, aOR 1.21, 95% CI 1.14 to 1.29
• Wound infection -- 7.4% versus 4.6%, aOR 1.24, 95% CI 1.20 to 1.28
• Pneumonia -- 3.7% versus 1.5%, aOR 1.17, 95% CI 1.12 to 1.22

Investigators performed analyses limited to patients who had hyponatremia documented ≤2 weeks before surgery, ≤1 week before surgery, and the same day as surgery. The adjusted odds ratios for 30-day mortality remained largely unchanged (aOR 1.35 to aOR 1.42).

The increased mortality risk persisted across multiple subgroups, but especially patients undergoing non-emergency surgery (aOR 1.59, P<0.001) and patients in ASA class 1 or 2 (aOR 1.93, P<0.001).

Limitations of the study included the potential for unmeasured confounders due to the study's observational nature, an inability to report on complications beyond 30 days, a lack of individual patient identifiers in the database raising the possibility of patients appearing more than once, and the absence of clinical data such as medication use.

Comment