Know when to report postoperative complications

Coding post-operative complications can be—well, complicated. Having a cooperative relationship with your physicians and improving perioperative documentation are two strategies that can help clear up murky coding dilemmas and reduce the need for queries.

For example, a patient is admitted for an elective colon resection for colon cancer. The physician documents a postoperative ileus, but then on the day of discharge states “expected postoperative ileus has been resolved.” Should the coder report code 997.4 (digestive system complications)?

ICD-9-CM Official Guidelines for Coding and Reporting states that reporting additional diagnoses must meet at least one of the following criteria:

  • Clinically evaluated
  • Diagnostically tested
  • Therapeutically treated
  • Results in increased length of stay
  • Results in increased nursing care or monitoring

In addition, the Guidelines state that you should not report signs and symptoms integral to a disease unless otherwise required by coding conventions.

Understand what qualifies as a complication

What defines a surgical complication? Coding Clinic, third quarter, 2009, p. 5, outlines three requirements.

First, the condition must be related to the procedure. “Just because I have an ileus after surgery doesn’t necessarily mean it’s due to the procedure,” said James S. Kennedy, MD, CCS, managing director of FTI Healthcare’s clinical documentation improvement (CDI) practice based in Brentwood, TN and Atlanta. “It could be due to a medication such as morphine or an anticholinergic. It could be due to an autonomic neuropathy. Physicians do not always document these, even though they know it in their head.”

Second, the condition should not be routinely expected or integral to the procedure. For example, an ileus for a short period of time is integral or expected with most bowel surgery, said Kennedy, who spoke during HCPro’s October 20 audio conference, “Top Five Coding and Documentation Challenges: Speak the Same Language as Physicians.” Likewise, impotence is routinely expected with a prostatectomy and postoperative respiratory failure is routinely encountered the first 24 hours after cardiac bypass surgery.

Finally, providers must document the condition as a complication, unless there is clear guidance from ICD-9-CM or Coding Clinic to code it as such. For example, because pneumothorax is a known risk associated with most thoracic surgeries, it would be inappropriate to assign an additional code for the iatrogenic pneumothorax based on an x-ray finding alone without physician concurrence, according to Coding Clinic, third quarter, 2003, p. 19.

Rely strictly on documented complications

Certainly attending physicians may have their own opinions of what qualifies as a complication after a particular procedure, but consider the following examples of postoperative complications:

  • Wound infections
  • Coronary artery bypass graft thrombosis
  • Infected joint prosthesis
  • Pulmonary emboli within one week of surgery

While these are not routinely expected or integral to surgical procedures, coders may not assume that they are complications unless supported by physician documentation, ICD-9-CM conventions, or Coding Clinic advice.

“In my opinion, Coding Clinic, third quarter, 2009, p. 5, is the most important Coding Clinic that discusses complications,” Kennedy said. “I’m not discounting the others, but this is probably the latest and greatest advice as to when conditions should be complications.”

This particular Coding Clinic explains that only physicians—not coders—can determine whether something that occurred during surgery is a complication or an expected outcome, Kennedy said. And when the documentation isn’t clear, coders need to query the physician.

“This is a sensitive subject when it comes to queries for the physicians because you say the word ‘complications,’ and of course, the physician gets a little bit nervous,” said Margi Brown, RHIA, CCS, CCS-P, CPC, CCDS, an independent consultant in Orlando and adjunct HCPro Boot Camp instructor. “But again, is it expected? And I think that’s one of the key words.”

Through education or a CDI program, coders and physicians can better understand the significance of the term “expected” in documentation. “I’ve actually seen the term ‘expected ileus’ documented more often, where it definitely helps the coders,” said Brown, who also spoke during the HCPro audio conference.

This brings us back to the original question posted at the beginning of this article. The ICD-9-CM Manual contains a default code for postoperative ileus—997.4. However, should you still report this code when the physician documents “expected postoperative ileus”? The answer may depend on facility policy or a physician query, Kennedy said.

But in this case, Kennedy believes only one code, 153.4 (malignant neoplasm of cecum), is appropriate because the physician stated that the postoperative ileus was expected and not due to medication. Therefore, the coder would not add code 997.4 nor 560.1 (paralytic ileus), given that the ileus did not meet the criteria for an additional diagnosis.

A code for ileus should not be added unless it meets the definition for additional diagnosis that affect patient care in terms of requiring clinical evaluation, therapeutic treatment, diagnosis procedure, extended length of hospital stay, or increased nursing care and/or monitoring. If there is any question, a query should be entered.

Take note of documented causes of conditions

Consider a case that involves a drug-induced ileus. Coding Clinic, January–February, 1987, pp. 13–14, included the following question and answer:

Question: How is spinal anesthesia-induced adynamic ileus of the colon coded? The anesthetic was tetracaine hydrochloride. The ileus occurred postoperatively, but the physician states it was not due to any intraoperative dissection.

Answer: Spinal anesthesia-induced adynamic ileus without mention of any error in dosage administration is coded:

• 560.1 (paralytic ileus)
• E938.7 (adverse effect of tetracaine administered for spinal anesthesia)

Note that in this case, the physician was clear that the ileus was due to the drug, not the surgery. So it’s important to identify when there is a documented postoperative ileus whether it was expected or whether there was another cause, Kennedy said.

Be aware of POA implications

Present on admission (POA) indicators can also be a factor in determining what qualifies as a complication. Whether or not a physician clearly states a condition is a complication, 3M’s Potentially Preventable Complications algorithm (used in Florida and Maryland) designate that some conditions with a POA indicator of N (no) or U (uncertain) deem them to be complications.

“The bottom line is we must establish the relationship,” Kennedy said. “Query, query, query. Interaction with the medical staff as to what is a complication and being careful not to underreport complications is very important.”

Editor’s note: E-mail your questions to Managing Editor Doreen V. Bentley, CPC-A, at To learn more about other conditions, such as acute kidney injury and altered mental status, for which frequent coding errors occur due to insufficient documentation, visit HCPro’s Healthcare Marketplace to purchase audio conference, “Top Five Coding and Documentation Challenges: Speak the Same Language as Physicians.”

Take this Quiz Earn 0.5 credit from AAPC and 1 credit from AHIMA.

Read the following articles prior to taking this quiz: