Anesthesia Coding Alert

Translation: Your anesthesiologist or pain management specialist should assign the most precise ICD-9 code to a service. You cannot justify a service with a four-digit diagnosis code when carriers or ICD-9 requires a more specific five-digit code to describe the patient's condition.

"Using the fourth or fifth digit when it is required -- or just when you do have that information -- is an important concept to follow," says Karen Marsh, RN, MSN, president of Kare-Med Consulting in Jensen Beach, Fla. Make sure you review the entire record when determining the specific reasons for the encounter and the conditions the physician treated, she says.

"Using a fourth or fifth digit for your diagnosis in some instances justifies the medical necessity for using an anesthesiologist to sedate patients," says Emma LeGrand, CCP, CCS, coding supervisor for New Jersey Anesthesia/Health Network Management in Florham Park.

Example: A gastrointestinal (GI) specialist evaluates a patient due to complaints of chronic abdominal pain. The physician decides to perform an esophagogastroduodenoscopy (EGD) and calls the anesthesiologist to sedate the patient because of anxiety (300.00, Anxiety state, unspecified).

The test is positive for irritant gastritis with hemorrhage (535.41, Other specified gastritis; with hemorrhage).

Caveat: If you submit the claim without the fifth digit that identifies the hemorrhage or if you submit it with the wrong fifth digit such as 535.40 (... without mention of hemorrhage), LeGrand says, you risk losing reimbursement for your anesthesiologist's services.

Code 300.00 detail: You can report the patient's anxiety as a secondary code, but it's not usually enough to warrant medical necessity for an anesthesiologist. Some carriers might accept it as justification, however, so check your guidelines.

Tip: Follow your physician's documentation when selecting diagnoses. If the medical record does not allow you to code to the required level of specificity, check with the reporting physician for guidance.

Avoid "rule outs": ICD-9 coding guidelines state that you should not report "rule-out" diagnoses in the outpatient setting.

Here's why: You'll avoid labeling the patient with an unconfirmed diagnosis -- which is inaccurate coding. Plus, by coding the presenting signs and symptoms, your physician will still get paid for his services, even if he cannot establish a definitive diagnosis.

"Look to see if the physician has given the patient a definitive diagnosis," says Denae M. Merrill, CPC, coder for Covenant MSO in Saginaw, Mich. "'Rule out,' 'suspected,' 'probable' or 'questionable' are not codable. If there is no definitive diagnosis given, look for any signs or symptoms that the patient has been having."

Review: CMS outpatient service guidelines explicitly state that practices should not use the condition being ruled out as the diagnosis. Instead, "code the condition(s) to the highest degree of certainty for that encounter/visit such as symptoms, signs, abnormal test results."

Pointer: Talk to your physicians about how important accuracy with their terms is. Tell the physician that if she can come to a definite conclusion about the patient's diagnosis, she needs to state this in her dictation so you may choose the best code.

Example: Many carriers only allow reimbursement for anesthesia during endoscopy procedures if you submit V58.83 (Encounter for therapeutic drug monitoring [use only for anesthesia utilizing intravenous propofol or a paralytic agent]).

Example: In the past, New York/New Jersey Medicare carriers paid for postoperative pain management epidurals only when you submitted V58.49 (Other specified aftercare following surgery). Beginning in 2007, however, the carrier approved using some of the new diagnoses for acute pain, including 338.11 (Acute pain due to trauma), 338.12 (Acute post-thoracotomy pain) and 338.18 (Other acute postoperative pain). The same policy states, "ICD-9-CM codes 958.8 (Other early complications of trauma) and V58.49 are no longer acceptable for reimbursement, as of 12/31/2006."