WHEN IT COMES TO COVID TESTS, Naznin Jamal, MD, says there are two she would like to be able to use in her hospital: a sputum test for intubated patients and a urine test for everyone else.
A urine test “would be so easily obtained,” points out Dr. Jamal, medical director of hospital medicine at Jefferson Regional Medical Center in Pine Bluff, Ark., an hour outside Little Rock. Plus, a urine test would bypass the problems with nasopharyngeal testing, including the variability of collectors’ experience and expertise and the lower-than-ideal sensitivity.
As for testing strategies, Dr. Jamal ideally would like to screen every patient in the ED. But given her hospital’s testing constraints, she can’t screen everyone who presents.
And because neither rapid sputum nor urine tests for covid currently exists, she and her 10-physician group are left “doing the best we can with what we have.” Although Dr. Jamal and her colleagues now have more testing options than just a few months ago, they still need to decide which tests to send out and wait for results vs. which to do in-house—and which patients to not test at all.
“Having access to enough tests would make our jobs easier.”
~ Naznin Jamal, MD
Jefferson Regional Medical Center
“Having access to enough tests would make our jobs easier,” she says. “By knowing within an hour which patients have actual infection, I would be able to select appropriate treatment and safely allocate limited hospital resources.” That would also, she adds, “help me understand the complexity of patients’ symptoms. This is not just a respiratory disease.”
Across the country, hospitals are forming alliances to share testing resources and—when they can—bringing new testing platforms onboard. But that doesn’t mean that many hospitals aren’t taking very conservative approaches to testing, to safeguard supplies that may be as scarce as they are essential.
Since the pandemic’s beginning, public health experts have warned that testing is the linchpin that every other aspect of care hinges upon. In hospitals, hospitalists have learned that a lack of testing—or unconscionably long turnaround times—fuel a cascade of other problems and shortages, including wasted PPE and negative pressure rooms being used when they may not be needed.
But in terms of turnaround times and testing options for hospitals, things have certainly improved since the pandemic’s Dark Ages back in March and April.
Namrata Singhania, MD, a hospitalist in Columbus, Ohio, says she and her colleagues, who are part of the Sound Physicians group there, used to have to send tests to the CDC and wait between seven and 10 days for results.
Now, says Dr. Singhania, while her hospital still sends some tests out to commercial labs, she gets results within 24 hours. Just as importantly, the hospital brought in Abbott equipment for its own in-house rapid test.
“With Abbott, we get results now within 40 or 45 minutes,” she points out. “The test isn’t 100% sensitive, but it’s made everyone’s life so easy.”
“We’ve come up with an internal triage based on the different turnaround times of different tests.”
~ Joshua Lenchus, DO
Broward Health Medical Center
In New Orleans, Geraldine Menard, MD, section chief of general internal medicine and geriatrics for University Medical Center New Orleans, likewise used to wait a week to 10 days for results to come back from the state health department. (At the time, the health department was the sole entity in that state processing tests.) Back then, Dr. Menard notes, the EDs at both University Medical Center and Tulane Medical Center—she practices at both—were flooded with infected patients from nursing homes.
“The doctors were doing all the paperwork and actually swabbing the patients with suspected covid,” Dr. Menard recalls. “And if patients coming in had been tested out in the community, we’d be on the phone to the health department, trying to chase down results. It was all very random.”
Since then, says Dr. Menard, both academic centers have brought all their covid testing in-house. “Tulane runs Abbott and Roche,” she says. “University Medical Center runs Abbott, Roche and Cepheid.”
Advantages of scale
While Dr. Menard’s two academic centers in New Orleans do all in-house testing, Banner Health in Arizona does the opposite: It sends all its covid tests out. But it is sending tests out to the state’s largest lab, Sonora Quest, a joint venture for which Banner—as the state’s largest health care provider—is the majority owner.
“Right now, Sonora Quest can do thousands of tests a day,” says Syed Jafri, MD, Banner’s physician executive for hospital medicine. For Banner inpatients, Dr. Jafri adds, turnaround time for results is less than 24 hours. “Capacity will continue to increase as additional testing lines are added until we reach 60,000 tests per day. That will improve Arizona’s readiness should the state experience a second wave of covid resurgence.”
As part of their joint venture, Banner and Sonora Quest have invested in a suite of different testing platforms. “We brought on Roche, Hologic and PerkinElmer,” says Dr. Jafri. “Multiple platforms allow us to maneuver around reagent shortages. It’s like an investment portfolio; you want some diversification.”
Joshua Lenchus, DO, agrees that diverse testing options makes sense. Dr. Lenchus, a former hospitalist who is chief medical officer of Broward Health Medical Center, the 700-bed Ft. Lauderdale hospital that’s the flagship of a four-hospital system, finds that “the more testing platforms you have the ability to use, the more possibilities you have.”
“It’s like an investment portfolio; you want some diversification.”
~ Syed Jafri, MD
The medical center was fortunate, he points out, to be in early talks with Abbott in mid-March. “We had really been waiting to pivot from Quest,” referring to the commercial insurer that has been intermittently overwhelmed with testing volumes. With an Abbott platform onboard, the medical center has since brought in other in-house PCR-based platforms: Cepheid’s GeneXpert and Hologic’s Panther Fusion.
The Abbott test has the fastest turnaround time of about 15 minutes, he explains. “But that platform tests only one specimen at a time, so you can process only 96 tests a day.” His health system processes more than 100 tests a day.”We’ve come up with an internal triage based on the different turnaround times of different tests,” he adds. In choosing which test to order, he also takes varying supply levels into consideration. “If I need a result within an hour, the Abbott or Cepheid provides that capability, but the Panther does not.”
Instead, he uses the Panther platform for next-day results. Or if patients are being discharged home from the ED and he’s sure they will self-quarantine until they get test results, “I could use Quest where the turnaround time could be several days.” The medical center also uses Quest to process the tests of patients coming in pre-procedure.
Another advantage of having multiple in-house platforms: If patients’ initial test in the hospital is negative but there’s still high clinical suspicion, the physicians at Broward Health treat them as patients under investigation and admit them to an isolation room. Then, “we test them again using a different platform.”
Rapid testing for high-risk patients
Smaller hospitals may not have that many testing options. In Pine Bluff, Ark., Dr. Jamal says she has for months relied on two testing methods: sending tests out to Quest—with turnaround times down from five days to under 48 hours— or running Cepheid rapid in-house testing, with results in 45 minutes to one hour.
“We get results now within 40 or 45 minutes. It’s made everyone’s life so easy.”
~ Namrata Singhania, MD
The dilemma: The in-house test, she notes, “is incredibly difficult to obtain, and we are not guaranteed supply. That’s true across the country.”
She and her colleagues recently gained access to an in-house antigen test (Sofia) that delivers results even faster than the Cepheid (15-20 minutes). That new option is helping them get through the Cepheid supply shortages. To stretch their in-house Cepheid supply even further, Dr. Jamal helped develop criteria for when to use that platform.
They’ve reserved the rapid in-house test for patients with a moderate to high suspicion of having covid. “With fast results,” Dr. Jamal says, “we can start treating them earlier.” In addition to patients at moderate to high suspicion, her hospital uses rapid in-house testing for all psychiatry inpatients, women in active labor, anyone undergoing emergency surgery or an aerosolizing procedure, and patients with STEMIs.
“They need to know to totally clean the cath lab after such a procedure,” Dr. Jamal points out. But for patients with low to moderate suspicion of covid, she and her colleagues send a test out to Quest instead.
At Saratoga Hospital in Saratoga Springs, N.Y., a 171-bed facility, physicians can send some covid samples to Albany Med Health System, their regional academic partner, to test.
But like the doctors at Jefferson Regional, they mainly rely on commercial send-outs (Saratoga uses LabCorp) and Cepheid for rapid-in house testing. And like Dr. Jamar, David Mastrianni, MD, senior vice president of the hospital medical group at Saratoga Hospital, says he views the hospital’s supply of Cepheid test cartridges as “very precious commodities.”
“For people we consider high risk, we can wait between 24 and 48 hours for their results to come back.”
~ David Mastrianni, MD
But while the Cepheid platform is his hospital’s “workhorse,” supplies are about to become even more scarce. “We just got an announcement that the rest of our summer supplies will be cut 25%,” says Dr. Mastrianni. “The company is retooling its manufacturing so that, in the fall, the cartridges will be able to test for covid, flu and RSV.”
Pooled testing for low-risk patients
To conserve their Cepheid stores, Dr. Mastrianni and his colleagues began implementing a pooled testing strategy in March. Although the Cepheid system is designed to burn one cartridge for each patient, they pooled three samples in each cartridge.
The strategy worked. Out of close to 200 cartridges used for more than 500 samples, only four came back positive and needed additional cartridges used for those individual patients. He and his colleagues published their results online in July in the Journal of Hospital Medicine.
The success of that pooling strategy depends on two key components, Dr. Mastrianni points out. First, Saratoga uses its rapid in-house test only for those patients deemed to be low risk.
“For people we consider high risk, we can wait between 24 and 48 hours for their results to come back,” he says. “We classify them as patients under investigation until then and treat them as if they’re positive.”
The second key component: a very low transmission rate in the community. According to Dr. Mastrianni, that rate has remained under 2%, even when his hospital was accepting critically ill covid patients from New York City during the crisis there. “In a hotspot, this isn’t the right technique,” he points out. “You’re going to infect a lot of pools.”
“Even if they’re discharged from the ED, patients are tested.”
~ Geraldine Menard, MD
University Medical Center New Orleans
The pilot study was so successful that the hospital has since pushed the number of samples it pools in each Cepheid cartridge from three to five. (See “Pooled testing: Pay attention to logistics.”) Out of roughly 3,300 samples tested to date at the hospital, “we’ve saved 2,300 cartridges.”
Pooling low-risk samples has also allowed Saratoga Hospital to test all patients being admitted, “and that was our goal,” says Dr. Mastrianni. Reaching that goal confers other advantages that are, he adds, “almost impossible to quantify, but important.”
For one, hospital staff feel more comfortable doing their jobs (and then going home to their families) knowing that all patients are tested. And patients out in the community, he believes, feel reassured that they can safely come to the hospital for non-covid care.
Has the hospital publicized the fact that it’s testing all patients? “We run some general messages that we’re trying to check everybody as best we can, without giving any guarantees,” Dr. Mastrianni says. “But in a community like ours, where the hospital is the area’s largest employer, when your staff knows it’s safe, the community does as well.”
At Broward Health Medical Center, says Dr. Lenchus, “all patients are tested, regardless of what they came in for.”
Likewise in New Orleans, Dr. Menard says that everyone coming into Tulane Medical Center or University Medical Center gets a test. “Even if they’re discharged from the ED, patients are tested. We use covid test results to triage patients into different areas of the ED.”
In Ohio, Dr. Singhania says she supports the idea of universal testing in the hospital. She was coauthor of a letter to the editor published in July in The American Journal of Medicine, which described one covid patient’s atypical presentation in the ED with only altered mental status.
“With such presentations, we may be seeing covid patients without knowing it,” she points out. But while universal screening may be ideal, “there are so many limitations to it,” she admits. “If we did universal screening, we’d be out of tests for those who really need them.”
That’s the same reason, says Dr. Jafri, that universal testing isn’t in place at Banner Health. All pre-procedural patients in Banner hospitals receive both PCR testing and IgG testing for antibodies. “These are patients being anesthetized, so we feel it’s best to have as much information on hand prior.”
While positivity rates among the general population in Arizona were running between 18% and 24% during the most recent surge, Banner has found a much lower positivity rate among asymptomatic patients in pre-procedure screening.
As supplies permit, Banner plans to expand testing. “Trauma patients, OB patients and inpatient surgery patients who aren’t having any covid-like symptoms—that will be our next iteration,” says Dr. Jafri. “But we’re not routinely screening all patients across the board. We need to be good stewards of resources to best serve our community.”
Pooled testing: Pay attention to logistics
IN SARATOGA SPRINGS, N.Y., Saratoga Hospital has (so far) successfully conserved supplies for its Cepheid rapid in-house testing platform by pooling test samples from low-risk patients. Saratoga is now using individual Cepheid cartridges to test five separate samples.
The community’s very low transmission rate makes that strategy feasible, says David Mastrianni, MD, Saratoga’s senior vice president of the hospital medical group. Another local hospital, which heard about Saratoga’s success, has been using the same approach to test pooled samples of patients undergoing surgeries.
Given that the strategy works among low-risk patients, how about using it in schools or to screen staff in nursing homes?
“I think pooling would be great in those settings,” says Dr. Mastrianni, “with this caveat: You have to be very careful with the logistics. How would the school system check everyone, turn results around and get results back?”
In the ED and hospital, for instance, “it’s easy to say, ‘Let’s wait for a batch of five.’ But then the ED is getting full, you have three patients waiting to be admitted and you hear an ambulance is coming in with two more.” In that case, he adds, “somebody has to make the call and say, ‘OK, we’re going to run this batch with less than five.’ ”
The clinician who makes that call, he notes, has to be able to balance out the needs of the ED and of the patients waiting for results vs. the scarcity of testing resources. And “that requires a lot of communication and support.” He estimates that such a call takes place in his hospital twice a week.
You also have to work out the logistics of what to do with the occasional positive pool. “How are you going to go back and redo all those samples?” he says. “That needs to be worked out by each hospital or setting doing pooling.”
To help make those decisions at his hospital, Dr. Mastrianni is one of five clinicians—dubbed covid physician advisors—who keep up with hospital statistics on test supplies. “One of us is always available,” he says, “if somebody has a question. That’s worked out nicely.”