CEO SUMMARY: In the report by the Pennsylvania Department of Health on the death of a patient at Lehigh Valley Hospital, it was noted that the clinical staff failed to notice discrepancies between results from point-of-care (POC) tests at the bedside and lab test results from blood serum run in the hospital’s central laboratory. Experts point out that this case is an example of a failure to notice these discrepancies, a failure to report on the discrepancies, and a failure in communication.
FOLLOWING THE DEATH OF A PATIENT in a Pennsylvania hospital earlier this year, state investigators determined that errors in protocols were a contributing factor. The state report on this case indicated the clinical staff failed to notice discrepancies between test results from point-of-care (POC) tests at the bedside and results from blood serum tests run in the hospital’s central laboratory.
For more than two decades, pathologists and clinical laboratory professionals have debated the accuracy and reliability of lab tests performed on point-of-care devices as compared to laboratory tests performed in central laboratories. This case provides a window on how and why differences in test results produced by the two methodologies have the potential to negatively affect patient care.
Using sources in the public record, the basic facts of this case appear to be that the patient was administered insulin to lower blood glucose levels, based on high blood glucose readings from tests performed with a point-of-care testing device. At the same time, in the patient’s medical record, there were blood glucose results that showed a lower number from lab tests performed in the hospital’s central laboratory.
In their report, state investigators determined that some members of the care team had noticed the discrepancies in the blood glucose results from the central lab and the POC tests, but that no effective steps to reconcile the differing blood glucose results were taken. (See sidebar on page 9 for a summary of the report produced by the Pennsylvania Department of Health.)
Patient Died in January
The patient was Sister Maria Angelita Soliman Quito, age 51, a native of the Philippines, who was a teacher at Immaculate Conception Catholic School in Pen Argyl, Pennsylvania. She was treated at Lehigh Valley Hospital Cedar Crest, in Allentown, Pennsylvania, and died on January 6, 2011.
Attorney Wayne Schaible of McCann, Schaible & Wall, LLC, in Philadelphia, represents Sister Angelita’s estate. “One of the issues in the case is whether the POC test strips for blood glucose testing were defective, but there are more important issues in this case,” Schaible said.
“We have not yet been given access to the POC strips or to the meter, but from our perspective, whether the strips or the meter were defective is irrelevant because, before you start pumping insulin into someone who has no history of diabetes, you should do a lab test,” he commented. “We were shocked to find that they did laboratory tests but they ignored the results of these tests.
“The patient had a kidney removed a number of days earlier but she was readmitted when she developed flu-like symptoms,” he continued. “They decided to use a point-of-care test for a blood glucose level and when they saw it was high, they put her on intravenous insulin.
POC vs. Core Lab Testing
“They should have looked at the central lab test results before they gave her insulin,” he noted. “The POC strips could be wrong for a number of reasons. They could be outdated or subject to storage issues. And the POC glucose meters could be wrong. They have to be calibrated regularly. But the larger point is that apparently they just ignored the central lab test.”
Department of Health Report
A report from the Pennsylvania Department of Health showed the patient’s blood sugar levels from blood serum tests done in the central laboratory were between 1 and 3 milligrams per deciliter (mg/dl).
THE DARK REPORT asked a recognized national expert in point-of-care testing to review the public information and comment on what lessons might be drawn from this case that would be useful to pathologists and clinical laboratory managers. Because of the ever-expanding use of POCT, the accuracy of POC test devices versus tests performed in central laboratories is a topic of high interest.
Further, because of pending litigation and because the public record about this case is not complete, THE DARK REPORT chose not to identify the expert. The views presented here represent the expert’s assessment of how protocols involving laboratory testing may not have been fully followed by the patient’s care team.
Articles published by the Allentown newspaper, The Morning Call, were one source of information about this case. After reviewing this information, the lab expert stated, “From the press coverage of this case, it appears that many issues were involved.
Low Glucose Levels
“The staff was relying on POC testing and it’s clear from the newspaper that the patient had low glucose levels according to the tests coming from the central laboratory,” commented the expert. “Keep in mind that central laboratories have critical-value call-back policies that are part of the required follow-through steps defined under recommendations of The Joint Commission (TJC).
“It appears that this patient had a critical value, meaning the main lab would have called a clinician on the patient’s floor, either a nurse or a physician,” explained the expert. “The central laboratory would have alerted the floor that there was a low level of blood glucose in this patient.
“The floor should have alerted the clinicians treating the patient,” he continued. “So, it’s not just that someone didn’t remember to pick up the note that the glucose test performed in the central laboratory showed low values. They might have been notified about the low values and not acted on that information. Alternatively, that information was not passed to the staff at the bedside.
“It appears there could have been a problem with the POC testing and with the lack of communication as well,” he added. “It’s impossible to know where the failure in communication occurred.
“There is one other important question about this case and that is whether there were lab values of 1 to 3 mg/dl,” he commented. “It’s almost physiologically impossible to get a lab value that low from a living patient.
“Such a critically low glucose value would have been called immediately to the floor by the staff in the central laboratory,” observed the expert. “That’s why this case shows that there was a break- down in many areas and that the staff in all hospitals and care settings can, and should, do a better job of communicating.
“In addition, there is an aspect of this case that laboratory professionals need to recognize regarding POC technology,” stated the expert. “POC glucose meters can do only what they are designed to do and that is provide estimates of the patient’s glucose. They give only a ball-park estimate of whether the patient’s glucose is high, low, or in the middle of the normal range.
“The death of this patient is a very unfortunate situation, and it is a lesson for every organization that uses POC testing,” noted the expert. “Whether test results come from a POCT device or the central laboratory, there is a need to be cautious. Laboratory professionals understand the limitations and capabilities of different test methodologies. If there’s anything we can learn from this case, it’s the need for better communication.”
Hospital Issues Statement
Most pathologists and lab administrators are familiar with the challenges of performing point-of-care testing in hospitals and health systems. One issue is the capabilities of the technology incorporated in the POC assay. It is often not as robust as an assay that is performed on a calibrated analyzer in the central laboratory.
The second issue is the knowledge, training, and operation of hospital staff who may perform point-of-care tests. It is not uncommon for these operators to overlook important steps in performing a point-of-care test. Such omissions can affect the quality and reliability of the POC test results in such instances.
After this incident in January, the hospital issued a statement which stated that, because of expected legal action, it would be inappropriate to comment except to say the following: “We were extremely saddened by our patient’s passing. All of our physicians, nurses, and staff are dedicated to carrying out our mission of healing, comforting, and caring for each patient. At the time it occurred, we reported the incident to the appropriate authorities and took immediate action to improve our processes and procedures.”
Pennsylvania State Department of Health
AN INVESTIGATIVE REPORT ISSUED BY THE PENNSYLVANIA DEPARTMENT OF HEALTH explains the steps clinicians took when caring for the patient who died in January as a result of a number of errors related to point-of-care testing at Lehigh Valley Hospital in Allentown, Pennsylvania.
“Review of point-of-care (POC) glucose monitoring testing and nursing documentation for MR1 [the patient’s medical record] dated January 3, 2011, revealed that patient blood glucose was elevated between 253-480 mg/dl (milligrams per deciliter),” the report said. “Further review of nursing and laboratory documentation revealed that the patient’s blood serum glucose was between 1-3 mg/dl at the same time. There was a discrepancy between the high results of the patient’s POC blood glucose tests and the low laboratory serum glucose results.
“Review of Intensive Care Unit (ICU) nursing documentation and Advanced Intensive Care Unit (AICU-telemedicine) nursing documentation for MR1 dated January 3, 2011, at 4:00 a.m. revealed that the AICU nurse was aware of the discrepancies between the laboratory blood serum and point-of-care glucose tests results,” the report said. “Review of MR1 revealed no documented evidence that the ICU or the AICU nurse communicated to the physician the discrepancies between the high POC testing versus the low laboratory blood glucose results.
“Review of the AICU telemedicine physician orders for MR1 dated January 3, 2011, at 6:06 a.m. revealed that an order was entered for Lantus (medication to lower blood glucose) 20 units, subcutaneous in the morning in addition to the insulin drip,” the report said.
“Review of physician documentation for MR1 revealed no documented evidence on January 3, 2011, from 7:00 p.m. to 7:00 a.m. of the patient’s progress notes from the AICU telemedicine physician who provided services to the patient… concerning the discrepancies between glucose testing results,” the report said.
“Review of physician documentation for MR1 dated January 3, 2011, revealed no documented evidence that the patient’s attending physician was notified of the patient’s condition by the AICU telemedicine physician,” the report said.
“Review of physician documentation in a neurology consultation for MR1 dated January 5, 2011, revealed that the patient was found in the morning of January 3, 2011, by the surgical team with fixed dilated pupils and was in an unresponsive coma,” the report said. “The neurologist’s impression for the patient in MR1 was documented as coma secondary to prolonged hypoglycemia that met brain death criteria.”
It Is Important to Understand the Limits of Point-of-Care Testing Technologies
POINT-OF-CARE TESTING is not a means that anyone—physicians, laboratorians, or nursing staff—should use when looking for accuracy of precise glucose levels,” stated one laboratory professional who is considered an expert in point-of- care (POC) tests and devices. “These devices are not high-precision laboratory instruments. They are just disposable technology to give a rough estimate of glucose or trends over time.
“Individual hospitals spend hundreds of thousands of dollars on precision lab instrumentation in the central laboratory to produce very precise traceable levels of glucose,” stated the expert. “Those results cannot be compared to a disposable point-of-care test performed on the patient floor.
“Having said that, millions of these POC tests are used each year without incident because people understand their limitations,” he continued. “Caregivers understand that they can use them as a guide. Then, if they see low values, they can rely on different technology, meaning the central lab blood serum test results, as a fall back.
“There are limitations to all lab testing methodologies, and it’s our job as laboratorians to educate the people ordering the tests and interpreting the results so that they understand those limitations,” concluded the POCT expert. “Sometimes the POCT results can be compromised by drugs and hemolysis of the sample.