A Good Procedure for Procedures

by Ann S. Lofsky, MD, Anesthesia Consultant and Governor Emeritus

Interventional physicians are often asked to perform procedures on patients whose medical care is managed largely by others. They may be consulted because of their abilities to perform specialized invasive diagnostic tests or therapeutic maneuvers, and may interact with patients primarily during this time. Much of their attention is rightly focused on exercising skillful technique. Most doctors are careful to later document detailed descriptions of the medical procedures they do. Their attention may not be as focused, however, on the performance or documentation of preprocedural patient evaluations and postprocedure follow-ups, although, as the following case illustrates, these actions may be just as important as the performance of the procedure itself in ensuring good patient care.

A 62-year-old female with chronic hypertension and a history of alcohol abuse complained of right upper quadrant pain of several months’ duration. Physical examination and routine laboratory work were within normal limits, with the exception of mildly elevated liver enzymes. A CT scan revealed multiple small mass lesions in the right lobe of her liver possibly consistent with metastatic disease. After discussing the options with her internist, the patient was scheduled for a CT-guided core liver biopsy by our insured radiologist. The insured met the woman for the first time on the morning of the procedure in the hospital radiology department. He had the history and physical from the internist available in the chart. The radiologist discussed with the patient what the biopsy would entail and listed the possible complications—including bleeding, infection, or damage to the organ itself—and the patient signed an informed consent to that effect.

The biopsy procedure was performed uneventfully using an 18-gauge needle. Three samples of the right lobe were obtained without difficulty, and a post-biopsy film taken after removal of the needle was negative. As was the hospital routine, the patient was then taken to the Ambulatory Care Center for monitoring, where her vital signs were checked for 30 to 40 minutes. While getting dressed afterward, the patient told a nurse she was feeling a bit “shaky.” The nurse asked whether she would like to be evaluated in the Urgent Care Center nearby, but the patient declined, stating that she would rather go home. The woman and her husband were told to call the hospital should any problems develop. She phoned her internist later that afternoon requesting something for pain, and a prescription for Vicodin was called into the local pharmacy for her.

Did the Radiologist Meet the Standard of Care?
The following morning, the woman collapsed at home. Paramedics were called, and she was transported to the nearest emergency room with full CPR in progress. She was successfully resuscitated after multiple attempts at cardioversion, but she remained unconscious and ventilator-dependent. Initial laboratory testing in the ER revealed a hemoglobin of 5.2 and a hematocrit of 17.1. Coagulation tests showed a PT greater than 100 and a PTT greater than 200, which was as high as that hospital laboratory recorded. A neurologist ultimately diagnosed the woman with severe hypoxic encephalopathy, and, at the family’s request, she was placed on DNR status. She expired shortly afterward. An autopsy revealed a very large amount of clotted blood in her peritoneal cavity and a small pancreatic carcinoma with liver metastases. The patient’s family filed suit, alleging negligent preoperative evaluation and monitoring for a liver biopsy.

Expert reviewers criticized the insured for the absence of any clotting studies prior to the liver biopsy. The plaintiffs’ experts argued that the standard of care requires that clotting studies be obtained prior to going forward with any biopsy procedure likely to cause significant bleeding—especially in the presence of impaired hepatic function. The liver biopsy specimens were ultimately read as showing significant cirrhosis, which likely decreased hepatic synthesis of clotting factors and caused a hypercoagulable state. The insured stated in his deposition that he did not order a PT or PTT before the biopsy because he was under the impression that these tests had already been performed and were within normal limits. He said he recalled having been informed of this by the nurse. The nurse did not recall this, however, and no pre-biopsy clotting test results were found in the patient’s medical records, either at the hospital or in the internist’s office.

Additionally, reviewers were highly critical of the fact that there were no recorded vital signs following the procedure, and, in fact, no documentation of any kind for the time the patient spent in the ambulatory care area. She had never been formally admitted to this area of the hospital, and it was not the protocol at that time for nurses to document on patients who were simply being “observed.” Some reviewers commented that the period of observation was too short and should have extended for several hours, during which time some instability might have been noticed. The insured was also criticized for not supplying the patient with any written instructions on what to do if significant problems developed. When the patient complained of pain, the plaintiffs’ experts argued that most radiologists would have asked her to return to the hospital to check for evidence of bleeding.

While the patient’s life expectancy would very likely have been limited by her underlying pancreatic carcinoma, it was difficult to argue that the insured’s liver biopsy was not the proximate cause of her death and that obtaining baseline clotting studies and/or conducting better monitoring and follow-up after the procedure may have prevented her from expiring at that time. Because of difficulties finding expert support for the insured and the concerns about the poor documentation in several areas, the radiologist and The Doctors Company agreed to settle this case within his policy limits.

What Can Be Learned from This Claim?
A major problem in this case was the absence of clotting studies before the biopsy. The insured traditionally relied on nurses to check laboratory tests for him and to tell him about any concerns. As this claim illustrates, however, the ultimate responsibility for evaluating whether patients are acceptable candidates for the procedures they have scheduled remains that of the physician performing them. The delegation to others of a crucial task such as this one can have disastrous results.

The hospital’s protocol for handling patients after invasive procedures was never formalized, and had simply developed over time. In deposition, the insured was forced to admit that he was not aware of exactly how long his patients were observed and did not have any written discharge instructions or criteria. However, his responsibility to the patient included ensuring adequate post-procedure care, and having written guidelines would have gone a long way toward accomplishing that. The radiologist claimed that, as he always does, he personally monitored this patient with an automatic blood pressure cuff, ran an EKG strip, and performed a physical exam while the patient was in the ambulatory care area. Unfortunately, it was never his habit to document these actions in the medical record. And, as risk management/patient safety experts are fond of saying (only slightly tongue-in-cheek): “If you didn’t write it down, it didn’t happen.” While that may sound extreme, in medical malpractice litigation, it may well be what a jury will conclude. If it is important enough to do, write it down.

Communication was felt to be a problem in a number of areas. When the patient complained of feeling unwell in the ambulatory care area, the nurse rightly suggested that she remain for further evaluation. Yet when the patient said she wanted to go home, the nurse simply let her go without notifying a physician (or even documenting it in the patient’s chart). Had the radiologist been made aware that the woman had complaints, it might have been a clue for him to reevaluate her for bleeding prior to discharge.

Reviewers also commented on the apparently poor communication between the internist’s office and the radiology department. The internist could have been advised at the time of scheduling the procedure what laboratory tests would be required, and when the patient phoned complaining of pain, the internist should have notified the treating physician. Rather than focusing only on their individual roles, if the physicians and nurses had discussed the broader picture of this patient’s care together, the ultimate outcome in this case might have been averted.