Picking a path for PTH testing
Parathyroidectomy has an excellent success rate, with a 95 percent chance
of curing a patient with hyperparathyroidism. But the operation as it is performed
at many hospitals has an element of unpredictability: It may be complete, but
a critical step remains before the anesthetized patient is ready to be awakened.
The laboratory conducts a single test to determine whether all the diseased glands
have been excised, or whether the surgeon needs to explore further.
That test is an intraoperative assay of intact parathyroid hormone, or PTH,
and it must be performed rapidly. In fact, the first commercially available
rapid PTH assay was designed as a point-of-care test only about six years ago,
and many hospitals, when they began performing rapid PTH assays, did them right
in the operating suite. But in an unusual turnabout, despite the continuing
availability of portable instruments, at a growing number of hospitals rapid
PTH testing is migrating back to the central laboratory.
The College’s Point-of-Care Testing Committee discovered the trend in a 2001
poll of 370 responding labs participating in the CAP Special Ligand Survey.
"Initially, intraoperative PTH was meant to be an operating room assay," says
committee junior member Alexis B. Carter, MD. "We were surprised to find that
71 percent of laboratories performed the test in a central laboratory, six percent
in satellite laboratories, and only 23 percent in operating suites." It’s an
interesting example, she says, of how the commercial availability of new assays
can drive some tests back into the central laboratory.
Intraoperative PTH testing is probably the most widely applied intraoperative
hormone measurement, with two main manufacturers in the United States. Nichols
Institute Diagnostics, which makes the Quick Intraoperative PTH assay, has contracts
with 80 U.S. hospital laboratories, and Diagnostic Products Corp. has an approximately
equal number of accounts for its Immulite Turbo Intact PTH assay. In the poll,
about 20 percent of laboratories that reported doing rapid PTH used other methods
such as the Roche Elecsys analyzer.
Roughly three in 10,000 persons are affected by primary hyperparathyroidism.
In the outpatient population, it’s one of the most common causes of hypercalcemia.
"Stones, bones, and groans" is the medical-school mnemonic that summarizes the
primary manifestations of hypercalcemia, which often leads to kidney stones,
accelerated bone loss, and gastrointestinal discomfort. But only about 15 percent
of patients with hypercalcemia have symptoms.
For patients with hyperparathyroidism, the only definitive therapy is parathyroidectomy,
a procedure that in the past has had a failure rate of up to five percent in
experienced hands. It’s complex, because the locations of the four small parathyroid
glands in the neck and mediastinum vary from patient to patient, and some patients
have extra glands.
"Typically, there are four," says Dr. Carter, who is a pathology resident
with the Brody School of Medicine at East Carolina University in Greenville,
NC. "However, three to as many as seven have been reported in a single patient."
In addition, it is difficult to tell adenomas from hyperplasia either by surgical
inspection or histologic examination.
At least one standard surgical textbook, in fact, still advocates the traditional
approach to parathyroid surgery: a bilateral neck dissection. In this procedure,
Dr. Carter says, "surgeons make a neck incision, explore both sides, and attempt
to identify all four of the parathyroid glands in their usual positions—in
and around the thyroid gland. If any are enlarged, they resect and submit them
for frozen section tissue confirmation."
"In most cases a good surgical pathologist can identify whether the specimen
is a parathyroid gland versus a lymph node or nodule of thyroid tissue," she
continues. "But the single biggest problem with parathyroid surgery is the possibility
of missing an ectopic hyperfunctioning parathyroid gland."
Once researchers discovered that hyperfunctioning parathyroid glands took
up a certain radioactive agent, physicians started doing preoperative localizing
scans, called technetium-99m sestamibi scans, on these patients. "The idea was
if you found only one enlarged gland by the scan, you wouldn’t have to do the
bilateral procedure," Dr. Carter explains.
"The reason that this is important is that whenever you operate in general,
especially in the neck, the scar tissue formation can be quite intense. If for
any reason someone has to go back in and re-operate, the scar tissue can obscure
the anatomy and cause significant difficulties. The chances of surgical failure
increase dramatically when operating for the second time, and there’s a much
higher risk of complications."
But the nuclear medicine scan isn’t perfect. Walter E. Pofahl, MD, an associate
professor of surgery at the Brody School of Medicine, says the test’s sensitivity
for a single adenoma is between 70 percent and 85 to 90 percent. That’s where
the rapid PTH assay comes in. Says Dr. Carter, "If the surgeon removes the preoperatively
localized gland and the patient’s PTH level drops dramatically, then the surgeon
may feel comfortable that all of the diseased tissue has been removed. The remainder
of the neck may be left untouched, which decreases patient complications."
"If the scan shows a single gland," she says, "the surgeon takes the patient
to the operating room and goes directly to that area." Before the surgeon performs
the incision, he or she draws a baseline rapid PTH assay, exposes the gland,
and removes it. "At that point the surgeon waits," Dr. Carter says. "At our
institution it’s 10 minutes, but in the published literature it varies from
five to 15 minutes. Then the surgeon draws another PTH level. If the second
level has decreased from the baseline level by 50 percent, then this is indicative
of complete removal of hyperfunctioning parathyroid tissue. The surgeon will
close the patient and end the procedure."
"If the PTH does not decrease by 50 percent, the surgeon may first send the
resected gland for a frozen section to confirm that parathyroid tissue was indeed
removed." Then, based on the results, he or she will start to explore for an
undetected enlarged gland.
Dr. Carter and Dr. Pofahl work at Pitt County Memorial Hospital, the tertiary
care medical center for a 29-county area in eastern North Carolina, and surgeons
there perform 30 to 50 parathyroidectomies per year. They use the Diagnostic
Products Corp. assay, and the test is conducted only in the central laboratory.
When Dr. Pofahl performs a parathyroidectomy, he removes the gland, then waits
10 minutes and draws a rapid PTH. "The DPC takes 14 minutes. The Nichols is
a little quicker; once it’s in the laboratory it takes 10 minutes," he says.
"During that period, the patient remains anesthetized," he continues. "We
go ahead and close the incision, because at that point if the second PTH level
falls below 50 percent of the baseline, or drops by half, then it shows we’ve
taken out the right one. It’s not too often that you need to go back" and continue
"I did one last week where the initial baseline was above 200 and the followup
was less than 100, so the operation was completed."
The assay is most helpful in allowing a minimally invasive parathyroidectomy,
he says, but it’s occasionally used to document results of a bilateral neck
procedure. "Occasionally before the operation the scan doesn’t show a single
gland. In that instance, we go ahead and do a full neck exploration and find
all four glands," Dr. Pofahl says.
In his experience, it takes about an hour for a minimally invasive parathyroidectomy,
and about 90 minutes for a whole-neck procedure. But he disagrees that there
are many more complications with whole-neck. "You’re trying to improve on an
operation with an exceptionally low complication rate and over a 95 to 98 percent
cure rate," he says. "It’s tough to improve on, and intraoperative testing is
more expensive. But if you can localize a single gland more quickly, you have
obvious cost savings."
Performing the test in the operating room has many advantages, Dr.
Carter says. On the plus side, "you have more rapid turnaround time, so the
surgeon can more quickly determine if he or she has to look for another enlarged
gland, and it practically eliminates the need for specimen transport and errors
associated with specimen accessioning in the central laboratory. The medical
technologist also gets some good experience with direct patient care."
Cost, however, is the downside. The pathology department at the University
of Texas Southwestern Medical Center in Dallas reported that the cost of 50
parathyroid gland surgeries over a five-year period was $760 per surgery for
intraoperative testing versus $360 for central laboratory testing.1
Dr. Carter views this as the best comparison on the subject because it includes
the costs of operating time, laboratory overhead, instrument and reagent costs,
technologist time, and other factors.
Paul G. Catrou, MD, the head of clinical chemistry at Pitt County Memorial
Hospital, says, "We saw the price [of the Nichols assay] back about two years
ago. We just kind of shrugged our shoulders and said the hospital probably wouldn’t
want to pay for it since it’s a lot cheaper to do it in the central lab using
a multiple-assay device. So we brought in the Immulite Turbo test instead."
That decision reflects a common pattern where an instrument dedicated to a
single test is weighed against an instrument that performs on a broader platform,
Dr. Carter says. While Nichols’ system is dedicated to rapid PTH, DPC’s Immulite
analyzer can do many other assays with different software, including regular
PTH as well as assays for other hormones, enzymes, and tumor markers, to name
"Our turnaround is typically 27 minutes, which includes transport and any
problems with the instrument," Dr. Catrou says. "One problem with doing the
assay so infrequently is that the system must fine-tune specimen transport from
the operating room to the central laboratory. Transport time can vary anywhere
from two minutes to 20 minutes." Dr. Carter says many hospitals are apparently
finding it more efficient to work at rapid transport of specimens and results
rather than move testing to the operating room.
At Pitt Memorial, the surgical suite is roughly 300 feet away from the laboratory.
According to laboratory protocol, Dr. Catrou says, "we meet a person from the
operating room halfway and should have a transport time of less than five minutes."
But, he notes, if the specimen had to come from the ambulatory surgery center,
which is half a mile away, that transport time would be impractical.
A key factor in intraoperative testing is the need to run quality controls—which
can take 17 to 30 minutes—as close to the time of the operation as possible.
"The laboratory has to have advance notice of the surgery, because for us to
have rapid turnaround time, controls have to be run in advance so the instrument
is ready," Dr. Carter says. "We have run into a situation at this hospital where
the surgeon wanted a rapid PTH intraoperatively and the laboratory didn’t know
about it beforehand."
Surgeons at the National Institutes of Health perform about one to
two parathyroidectomies per week, says Glen L. Hortin, MD, PhD, chief of clinical
chemistry at NIH’s Warren Magnuson Clinical Center and former chair of the CAP
Point-of-Care Testing Committee. Alan T. Remaley, MD, PhD, director of the immunoassay
laboratory at the NIH, calls the intraoperative PTH assay "almost indispensable"
for the NIH’s patient population. "The NIH treats a lot of patients with multiglandular
disease and patients with ectopic glands, who were not adequately treated in
prior surgeries. We use the assay in all of our cases, and it frequently leads
to changes in the surgical management of patients," Dr. Remaley says.
The NIH initially performed the assay in the operating room with the Nichols
cart assay, but it now performs the assay in the central laboratory. In collaboration
with Nichols Institute Diagnostics, the NIH developed a rapid PTH assay on the
Nichols Advantage analyzer, which is an automated immunoassay instrument. "Our
central laboratory is adjacent to the operating rooms, so it was more convenient
for us to do the assay on a standard analyzer in the main laboratory," Dr. Remaley
says. The decision on where it’s best to perform the assay depends on many factors
that are "likely to vary from institution to institution, such as staff availability,
complexity of the cases, and the specimen transport time from the operating
rooms to the central laboratory," Dr. Remaley says.
Point-of-care instruments can be simple where waived technology is involved,
Dr. Carter says. "But this is not a waived test. The instruments can be very
difficult to operate," and a qualified person is needed to perform the test.
Some states might actually require a certain certification level for such
a complex test, but even in general, says Dr. Hortin, "It’s more suited for
a medical technologist to run because it’s fairly technique-dependent and requires
handling of calibrators, significant knowledge of instruments, and appropriate
timing when moving the samples from mixing to measurement." A better level of
automation would be required to get away from needing a medical technologist,
Some studies have shown, however, that despite the high cost of rapid point-of-care
PTH assays, the net savings are substantial. Researchers at the Washington University
School of Medicine, for example, studied an experimental group of patients for
whom intraoperative rapid PTH assays were between $55 and $113. Comparing this
group with a historical control group, the researchers found significantly fewer
frozen sections, shorter hospital stays, greater use of local anesthesia, and
more unilateral rather than bilateral neck explorations.2
Improvements to the test could further increase its usefulness in the operating
"One issue that’s starting to raise its head in the PTH arena over the last
couple of years is molecular forms of PTH which are not the intact molecule
but which are detected by the PTH assay," says Dr. Catrou. PTH consists of an
84 amino acid polypeptide, and the surgeons’ main interest is measuring this
intact molecule. "But there are fragments of PTH we have been having difficulty
with for many years in assays," he says.
When PTH is secreted from the gland, he explains, it’s released as an intact
active molecule. "What happens once it’s out in the blood is that it’s fairly
rapidly degraded by enzymes and primarily excreted by the kidneys. It can degrade
into different fragments, none of which are biologically active. One of these
fragments has a very short aminoterminal truncation of the first six amino acids
"We’re calling the latest assay ’intact PTH,’ but it turns out that even the
Nichols and the Turbo assays will actually respond to large PTH fragments that
are not the complete molecule," he continues. "These fragments of PTH, including
the PTH(7-84) fragment, can markedly be increased in patients with renal failure
and cross-react with the assay. So if you get an answer of 100 picograms per
milliliter of hormone, it won’t all be the intact molecule, and it’s the intact
molecule we’re interested in."
"Say you have someone with renal failure with a lot of PTH(7-84) in the blood.
You might not see your 50 percent decrease when you take out that parathyroid
because the assay may be measuring those smaller fragments as well as the intact
Although in general the assays have done very well, Dr. Catrou says, "I’ve
seen up to 50 percent cross-reactivity, so in certain cases your assay is going
to be 50 percent incorrect." Getting a new assay that measures the entire molecule
from amino acids 1 to 84 is an area that is being researched.
In fact, Nichols Institute Diagnostics released in December 2001 a standard
PTH assay, called the Bio-Intact PTH(1-84), that recognizes only the intact
(1-84) PTH molecule. Dr. Remaley at the NIH is evaluating a rapid version of
this new assay on the Nichols Advantage analyzer for use during parathyroidectomies.
The poll attached to the CAP Special Ligand Survey found considerable variation
in testing practices for parathyroid surgery, and Dr. Hortin suggests that more
evaluation is needed to determine the optimal use of intraoperative PTH testing.
Of 355 laboratories reporting PTH testing results, 92 said they offer intraoperative
PTH testing, and 81 of those reported the percentage of time it was used. Forty
percent said it was used for every case, 31 percent use it for a majority of
cases, and in 30 percent of the laboratories it is used for selected cases.
However, PTH testing remains far and away the most common intraoperative hormone
measurement and may serve as a model for other tumors. Only four laboratories
responding to the CAP Survey poll indicated they perform intraoperative testing
for other hormones. The clinical laboratory at the NIH, for example, has developed
and now performs rapid immunoassays for insulinomas, ACTH-producing endocrine
tumors, and cortisol-producing tissue.
Over the next few years, intraoperative PTH testing practices could change
again, Dr. Hortin notes. "If immunoassay analyzers become smaller, cheaper,
and easier to operate, tests may migrate back into the operating room," he says.
"We saw from our survey that the types of commercially available products strongly
influence the site of testing."
1. Wians FH, Balko JA, Hau RM, et al. Intraoperative vs central
laboratory PTH testing during parathyroidectomy surgery. Lab Med. 2000;31:616-621.
2. Johnson LR, Doherty G, Lairmore T, et al. Evaluation of
the performance and clinical impact of a rapid intraoperative parathyroid hormone
assay in conjunction with preoperative imaging and concise parathyroidectomy.
Clin Chem. 2001;47:919-925.
Anne Paxton is a writer in Seattle.
The full results of the 2001 poll of laboratories participating in the
CAP Special Ligand Survey will be published in an upcoming issue of the Archives
of Pathology & Laboratory Medicine in an article written by Drs. Hortin and