Author Affiliations: Division of Nephrology, Department of Medicine (Dr Vassalotti), and Departments of Obstetrics, Gynecology, and Reproductive Science and Health Evidence and Policy (Dr DuPree), Mount Sinai Medical Center, New York, New York.
In this issue of the Archives, Leung and colleagues1 present an interesting and important epidemiological analysis of almost 1 million patients who underwent surgery between January 1, 2005, and December 31, 2010, using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP), a validated database with nearly 300 participating academic and community hospitals. They demonstrate that preoperative hyponatremia compared with a normal preoperative serum sodium level is significantly associated with a higher risk of perioperative 30-day mortality (adjusted odds ratio [aOR], 1.44) and increased morbidity (major coronary events [aOR, 1.21], wound infections [1.24], pneumonia [1.17], and approximately 1 additional adjusted median hospital day). Preoperative hyponatremia was common, defined as a serum sodium level less than 135 mEq/L (to convert to millimoles per liter, multiply by 1.0) within 90 days of surgery, occurring in 7.8% of patients, or 75 423 of 964 263. In addition, almost 89% of hyponatremia cases were mild, defined as a serum sodium level of 130 to 134 mEq/L. Although, as expected, the risk of death was directly related to the severity of the hyponatremia, somewhat surprisingly, nonemergency procedures and low-risk cases (as stratified by American Society of Anesthesia classes 1 and 2) were significantly associated with higher risk (aOR, 1.59 and 1.93, respectively). Increased risk in patients scheduled for elective surgery with mild and almost certainly asymptomatic hyponatremia presents an important opportunity for the internist performing preoperative medical consultation.
Hyponatremia is familiar to physicians as the most common electrolyte disorder, occurring in up to 15% to 30% of hospitalized patients,2 or approximately twice the proportion described in this ACS NSQIP study. The clinical importance of a low serum sodium level includes adverse outcomes associated with acute severe cases and more mild, asymptomatic cases. Also, overly rapid correction of serum sodium levels can result in osmotic demyelination. Hyponatremia is well recognized as an extremely sensitive marker of the severity of underlying comorbidities, such as the severity of heart failure. This is not a typical finding in healthy people. In most patients with hyponatremia, water retention is characterized by excess antidiuretic hormone (ADH) inappropriately suppressed for a variety of causes or due to nonosmotic hemodynamic stimuli from hypovolemia or edema-forming disorders with decreased effective arterial blood volume. The diagnostic approach is based on this pathogenesis, using serum osmolarity to first distinguish hypotonic, isotonic, and hypertonic forms of hyponatremia3,4 and then, in the group of patients with hypotonic hyponatremia, hypovolemic, euvolemic (eg, syndrome of inappropriate ADH secretion and psychogenic polydipsia), and hypervolemic (eg, severe heart failure and advanced liver failure) etiologies in the context of the history, examination, and urine sodium findings.3,4
The finding that preoperative hyponatremia is associated with increased postoperative mortality is not surprising given its underlying comorbidities. For surgical patients with hyponatremia, ADH stimulation can be drug induced or pain related. The observation that most hyponatremia described by Leung et al1 was mild and likely asymptomatic underscores the importance of recognition of the underlying etiology and comorbid condition(s). In the discussion of hyponatremia as a mediator or marker, Leung et al1 cite the recent study by Waikar et al5 as evidence that the association between hyponatremia and mortality is independent of any ADH-mediated mechanism. This finding is not generalizable for several reasons. First, the patients with dialysis-treated chronic kidney failure described by Waikar et al5 cannot develop hyponatremia as a result of vasopressin-mediated water retention in the absence of functioning kidneys. Second, vasopressin levels in the study by Waikar et al were not evaluated. Third, most patients who underwent surgery in this ACS NSQIP study were not treated with hemodialysis. More important, however, the issue is not that vasopressin or ADH mediates the adverse outcomes associated with hyponatremia. Although there is debate over the contribution of hyponatremia per se to morbidity and mortality risk, there is no dispute that the underlying diseases that contribute to the development of most cases of hyponatremia are major mediators of morbidity and mortality. This is the most likely pathogenic mechanism for how a low serum sodium level mediates the increased mortality and postoperative complications in the context of predominantly mild hyponatremia.
Is there anything treating physicians can do to reduce the operative risk associated with hyponatremia? First, although routine assessment of serum sodium levels preoperatively is not recommended, 79% of patients had preoperative serum sodium testing in this study. Obviously, the first question should be whether serum sodium levels should be tested. Studies evaluating the utility of “routine” tests have indicated that 60% of tests ordered are not required based on review of the history and physical examination findings.6 Furthermore, 2 retrospective studies7,8 of elective surgery found that only 0.22% and 1% of these tests, respectively, revealed abnormalities that might affect perioperative management. The preoperative evaluation should strive to determine whether the patient is in optimal health and whether the individual's condition could be improved before surgery. Previous hyponatremia and conditions commonly associated with hyponatremia are reasonable indications to perform serum sodium assessment in a subpopulation of preoperative patients.
If preoperative hyponatremia is established, recognition and potential risk stratification should be emphasized. Internist tools in common use for preoperative consultation assess risk of cardiovascular disease complications primarily and do not consider hyponatremia as a variable of interest.9 The association of preoperative hyponatremia with adverse perioperative outcomes raises a variety of key therapeutic questions. How should preoperative hyponatremia be treated? Can preoperative medical consultation and co-management attenuate the risk of hyponatremia or improve the serum sodium concentration in a timely and safe manner? Is there a role for vasopressin receptor antagonist therapy in preoperative hyponatremia? Severe cases require an immediate diagnostic evaluation and consideration for postponement of surgery to allow for correction, particularly if the case is elective. Mild hyponatremia, though, is the much more common situation, and, at a minimum, comorbidities require collaboration among specialties to ensure that the patient's condition is optimized before surgery. In addition, the possibility of undiagnosed comorbidities needs to be considered. Whether elective surgery should be postponed for the treatment of mild hyponatremia cannot be ascertained from this study, but the diagnosis should contribute to the informed consent process. The challenge lies in determining the next steps. Although the algorithm is relatively straightforward for treating hyponatremia, it is unclear how much this treatment should factor into a decision to proceed with elective surgical procedures. Water intoxication associated with excessive hypotonic fluid administration after surgery should clearly be avoided. Hypovolemia is particularly important to recognize clinically because correction with an isotonic saline solution is recommended and effective in most instances.4 The major principles of hyponatremia therapy should be applied, including recognition of the underlying cause as described previously herein and close serum sodium level monitoring to avoid overly rapid correction. An individualized approach considering hyponatremia in the context of the patient's comorbidities and the planned surgical procedure can be the only guide to the sequence of interventions. Weed10 emphasized that the impact of preoperative medical consultation may not be the consult recommendations alone but, more important, perioperative co-management with anesthesia and surgical colleagues.
Correspondence: Dr Vassalotti, Division of Nephrology, Department of Medicine, Mount Sinai Medical Center, One Gustave L. Levy Place, Box 1243, New York, NY 10029-6574 (email@example.com).
Published Online: September 10, 2012. doi:10.1001/2013.jamainternmed.2
Financial Disclosure: None reported.
Vassalotti JA, DuPree E. Preoperative HyponatremiaComment on “Preoperative Hyponatremia and Perioperative Complications”. Arch Intern Med. 2012;172(19):1482-1483. doi:10.1001/2013.jamainternmed.2