Health Information Management

Coding conundrums: Let’s clear the air about complications and septicemia

JustCoding News: Inpatient, April 14, 2010

Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

by Robert S. Gold, MD

I’d like to take this opportunity to clarify a few concepts that I think continue to confuse coders.

998.4—Foreign body accidentally left

During a February 2010 HCPro audio conference, someone questioned a statement I made while presenting a case study about a patient who underwent a cholecystectomy. The surgeon performed a sponge count at the end of the case and determined that one sponge was missing. A second count was performed; however, it occurred after the surgeon closed the incision. The surgical team requested an x-ray in the operating room (OR) while the patient was still anesthetized. The x-ray revealed the location of the sponge, and the surgeon reopened the wound to retrieve it. The surgeon then closed the incision.

I said during the audio conference that although an initial sponge count might show a discrepancy, second counts usually demonstrate that all sponges are identified. In this scenario, the incision was closed before completion of the second count, which revealed a discrepancy. I said coders shouldn’t assign code 998.4 (foreign body accidentally left in the wound) for this scenario. The caller corrected me, stating that Coding Clinic, first quarter 2009, states quite the opposite.

That certainly got me thinking.

I contacted Coding Clinic and learned that the guidance upon which the caller relied is based on algorithms and definitions from the National Quality Forum (NQF). I called the NQF and presented the case. The NQF said I was correct and that assigning code 998.4 for this scenario would be incorrect. The NQF encouraged me to speak with the physician advisor at the Agency for Healthcare Research and Quality (AHRQ). I did so and presented the case—as well as evidence and standards of quality related to sponge counts. The physician advisor referred me to the NQF definition of the “end of surgery,” which can occur in several possible ways. One is closure of the wound. The physician advisor agreed that this aspect of the definition is illogical.

Next, I spoke with the physician advisor for the quality measures for the American College of Surgeons (ACS). He agreed that coders should not report code 998.4 because the surgeon adhered to all quality measures to avoid accidentally leaving a foreign body in the wound. Also, the surgeon retrieved the sponge before the case concluded. However, the definition as presented by NQF could—and did—lead authorities astray. In the end, the NQF, AHRQ, and ACS agreed that the definition should be changed and that the advice published in Coding Clinic should similarly be corrected.

Until now, however, surgeons and surgical teams who have performed due diligence by counting sponges or other instruments have been dinged by coders who simply followed official guidance. Everyone had good intentions, but the data were based on a definition that required clarification and correction.

Verna Gibbs, MD, founder of NoThing Left Behind®, a national surgical patient safety project to prevent retained surgical items, also will be working to dispel the confusion. There are many procedures performed in various healthcare environments (e.g., ORs, radiology suites, physician offices, day surgery facilities) for which the rules and the definition of “end of case” differ. NoThing Left Behind identifies areas where mistakes can occur and provides strategies to avoid leaving foreign objects behind in a patient’s body.


Another issue I want to discuss pertains to definitions and coding guidance for septicemia. During this same February HCPro audio conference, I said that bacteremia pertains to the presence of bacteria in the bloodstream. I also said the condition generally presents with a classic group of symptoms (e.g., shaking, chills, rigors, high temperatures, high pulse rates, elevated white blood cell counts). I said viremia refers to the presence of viruses in the bloodstream and frequently presents with the same set of symptoms. I also said fungemia pertains to fungal pieces in the bloodstream and that it presents with the same symptoms. Septicemia traditionally refers to an infection of the bloodstream. It can be due to bacteria, viruses, or fungi in the bloodstream. But the mere presence of these organisms in the bloodstream may not imply septicemia.

Sepsis occurs when the body responds to injured or dead tissue related to an infection. It occurs through the mediation of chemicals released into the bloodstream by macrophages. A systemic inflammatory response also can occur without an infection. Calling this sepsis is inappropriate.
Septicemia can occur with or without a localized infection somewhere else in the body tissue. Sometimes it is an infection in the bloodstream along with an infection elsewhere in the body (e.g., subphrenic abscess, ascending cholangitis, or acute ascending lymphangitis). These patients have sepsis and an infection constantly seeding the bloodstream. The heart valves, artificial hips, and hernia meshes are at tremendous risk of secondary infection. Other conditions involve septicemia with no localized infection (e.g., botulism and anthrax).

The term “septicemia” may be appropriate if there is, indeed, infection of the bloodstream. The physician may further determine that the infection is due to bacteria (bacteremia), viruses (viremia), or fungi (fungemia). The problem is many physicians use the term “bacteremia” when there is truly infection in the bloodstream. They argue that septicemia is an obsolete term.

Ask physicians to specify when bacteremia is an incidental condition and perhaps related to recent dental work or a transrectal prostate biopsy that the patient underwent. The bacteremia also may be completely asymptomatic bacteria in the bloodstream possibly related to an infected vascular catheter or something treatable with antibiotics. In this situation, report code 790.7.

Coders should report code 038.9 in the following scenarios:

  • The physician documents that bacteremia is related to an infection somewhere in the body that is seeding the bloodstream. The patient may need a full 10-day course of treatment or even six months of antibacterial therapy to keep the bacteremia under control.
  • There truly is an identified infection of the bloodstream caused by bacteria.

Current coding advice refers coders to code 038.9 when bacteremia is due to sepsis. This is impossible because sepsis does not presuppose bacteria in the bloodstream at all.

Editor’s note: This article was originally published in the April issue of Briefings on Coding Compliance Strategies. Dr. Gold is CEO of DCBA, Inc., a consulting firm in Atlanta that provides physician-to-physician programs in clinical documentation improvement. Reach him by phone at 770/216-9691 or by e-mail at

Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

    Briefings on APCs
  • Briefings on APCs

    Worried about the complexities of the new rules under OPPS and APCs? Briefings on APCs helps you understand the new rules...

  • HIM Briefings

    Guiding Health Information Management professionals through the continuously changing field of medical records and toward a...

  • Briefings on Coding Compliance Strategies

    Submitting improper Medicare documentation can lead to denial of fees, payback, fines, and increased diligence from payers...

  • Briefings on HIPAA

    How can you minimize the impact of HIPAA? Subscribe to Briefings on HIPAA, your health information management resource for...

  • APCs Insider

    This HTML-based e-mail newsletter provides weekly tips and advice on the new ambulatory payment classifications regulations...

Most Popular