And the band neutrophil counts play on​

And that was just one physician. “I spend a lot of time thinking about this—how do we have this conversation on a large scale?” Dr. Vergara-Lluri says. As her interactions with that one colleague showed her, in some cases it’s a matter of changing hearts as well as minds. “There’s got to be a way to appeal to the emotional side or the logical side of a clinician’s brain to say, We should not be doing this. It’s a little shocking, to be honest. How are we still talking about this 30 years later?”

Tests, like facts, can be stubborn things. Dr. Vergara-Lluri compares band neutrophil counting to ESR—another test laboratories would like to drop. “But it lives on,” she says, as a favorite among rheumatologists. “I’ve personally given up on that one.”

Dr. Bhargava sees similarities to bleeding time. Though it was long known to lack sensitivity and specificity, “It took decades for people to finally stop doing it.”

It’s also possible that larger forces are keeping band counts in orbit.

The study noted that the majority of laboratories reported doing band counts with all their manual differentials. “So it wasn’t a clinical department holding out and saying, ‘I want that band count,’” Dr. Bhargava says.

A number of sepsis clinical assessment algorithms, such as the St. John sepsis protocol, incorporate band counting. It’s likely, says Dr. Bhargava, that clinical guidelines—“presumably formulated without the input of laboratorians, or made with old data”—have not been updated.

Clinicians may be unaware of the impact of inter- and intraobserver variability on these scores, Dr. Bhargava suggests. “You input a number, but it may not be indicating what you think it’s indicating. So it’s not simply that it’s labor-intensive and time-consuming, which it is, but it’s also not doing any good clinically.”

It’s even possible, she adds, that clinicians would be equally surprised by the survey results. She offers a charitable take on the matter. “Maybe the labs that continue to report bands just haven’t thought about it. It could be that they’re just historically doing this, and their own clinicians may not even be using it anymore.”

Tradition can indeed be a culprit, says coauthor Olga Pozdnyakova, MD, PhD, associate pathologist, Brigham and Women’s Hospital; medical director, Mass General Brigham flow cytometry integrated laboratory service; medical director, Brigham and Women’s hematology laboratory and Harbor Medical Physician Diagnostics Laboratory; and associate professor, Harvard Medical School.

“For pathologists, it’s sometimes hard to influence the clinical community,” Dr. Pozdnyakova says. If not quite a battle of the bands, making a change does mean pathologists will have to confront a persistent chorus: This is how it has been done traditionally. In her own experience, she says, some clinical specialists “do not believe me. Because it’s always been done like that.”

Nevertheless, the authors express hope that the Archives article will be the jolt they need to reach out to clinical teams. Dr. Pozdnyakova says, “Now we actually have data from the study that show counting of bands is not appropriate, and it is time for a change.”

About a half dozen years or so ago Brigham and Women’s stopped reporting bands separately; now, they’re counted together with segmented neutrophils. Before the change, the laboratory did not include bands in the absolute neutrophil count. Explains Dr. Pozdnyakova: “Bands were not considered to be mature neutrophils, which is incorrect. So in several instances we would report a falsely low neutrophil count.

“It does have significant implications if it’s not assessed properly,” she continues. Physicians don’t necessarily know the intricacies of what’s considered a mature versus a nonmature neutrophil. “All they see is a normal white blood cell count and a low absolute neutrophil count,” which can send them chasing after more testing.

Dr. Pozdnyakova instituted the change to combine bands and neutrophils at Brigham and Women’s, using an approach that set fire to the usual, play-nicely playbook.

Asked how she achieved what was essentially a bloodless coup, she laughs, then says, “Well, to tell you the truth, we just sent a broadcast saying that because bands are considered to be mature neutrophils, the absolute neutrophil count will now be reported together with segmented and band neutrophils.”

And the response? “No one has noticed,” she says. You can almost hear the smile in her voice. “That was the best way. No one had questions.”

This unilateral move was unusual, she concedes. The size of her institution may have helped. “We always complain that sometimes the bigger the institution is, the more difficult it is to make a change. But I think when you are 100 percent confident your decision is correct, it’s almost easier to make that change because you don’t have to go to every single person—because there are tens of thousands of us.” A smaller institution might require more personal outreach, she says.

Occasionally, Dr. Pozdnyakova says, she received an email asking about the reasons for the change. “And then you’ll explain it to them, and they’ll be perfectly fine with it.”

Neonatology is, perhaps unsurprisingly, the exception. “Unfortunately,” she says. “They’re still asking us to perform band counts. Because neonatologists still swear by bands. When it concerns little babies, it’s really difficult to make changes. But we can start with grownups.”

How hard will it be to turn that particular tide? “My husband is a neonatologist at [Massachusetts General Hospital], and I cannot convince him,” Dr. Pozdnyakova says with a laugh. “If I cannot convince him, then I will have trouble convincing other people.”

Band counting isn’t performed at UCSF. The practice was discontinued there before Dr. Bhargava arrived six years ago.

But in her previous position in Boston, she was part of an effort to make the change. Much of it was driven by presenting data to clinical groups, she says.

Neonatal ICU physicians were the most resistant, she recalls, in part because there are few other parameters they can use for this patient population. “For a while we allowed them as an exception, and they could specifically ask for a band count. And then we slowly helped them transition.”

Many NICU physicians were accustomed to using the I:T ratio, or immature to total neutrophils, which requires band enumeration. Those who learned this as part of their training, Dr. Bhargava says, also tended to be more resistant to change.

But persistence paid off. Likewise, for laboratories ready to make a change, Dr. Bhargava urges pathologists to reach out to key groups. “Tell them there has been decades of data and papers that tell us how bad the variability is in band counting.”

Does an alternative test exist? Ideally, it would perform well, have a fairly tight confidence interval, and isn’t costly. (“Band counting is none of those things,” Dr. Vergara-Lluri says with a laugh.) The Archives article alludes to several promising possibilities, including immature granulocytes, C-reactive protein, and procalcitonin, though the literature in this field needs to grow.

“None are perfect,” Dr. Bhargava says. “No one CBC parameter is great at predicting sepsis. The bottom line is, if bands are of low added value and so are IGs, pick the one that is more automated and less time-consuming.”

And while Dr. Pozdnyakova’s maverick approach may not work everywhere—not everyone feels comfortable trading a three-point turn for a U-turn into oncoming traffic and then hitting the gas—laboratories do need to tackle the problem. “If every one of us starts in their own institution, I think we’ll achieve great results,” Dr. Pozdnyakova says.

“It sounds like a long road ahead,” agrees Dr. Vergara-Lluri, “but I’m hopeful.”

Karen Titus is CAP TODAY contributing editor and co-managing editor.