The Pregnancy Lab

Is the qualitative serum pregnancy test obsolete?

I’ve written several times about qualitative hCG tests in this blog.  As a reminder, qualitative tests can be performed using urine or serum samples.  Urine tests can be performed close to the patient or even at home because the urine sample requires no special processing.  However, when serum is the test really can’t be performed at home or at the point-of-care because the blood sample has to be centrifuged to first obtain the serum and centrifugation is usually only performed in the clinical laboratory.  Notably, clinical labs are often able to do quantitative hCG testing on serum, too.

So, if a lab can do qualitative and quantitative hCG testing on serum, why not just offer one test instead of two?  In other words, might the qualitative test be considered obsolete?  My lab recently published a study that addressed that question.

To answer that question we surveyed several hundred doctors and the survey results revealed the following:

1. When requesting serum hCG tests, 49% of physicians preferred to order a qualitative rather than a quantitative test even though they believed quantitative tests were more accurate.
2. Physicians preferred qualitative tests because they believed that they received the test results faster.

However, when we examined the turnaround time data, that last point was not supported.  There are a few definitions of turnaround time to consider.  Doctors consider it to be the time it takes to get a result after the sample is collected while laboratorians consider it to be the time it takes to produce the result after they receive the sample.

By the lab’s definition, qualitative tests were performed more rapidly than quantitative tests but there were no differences using the doctors' definition of turnaround time.  That’s because the time it takes to transport the sample to the laboratory is known to contribute the most to delays in the total testing process.  So, although physicians believed they get results from qualitative tests more quickly, it doesn’t seem to be the case.

We also compared the analytical sensitivities of the two types of tests.  The qualitative test that we used had a claimed detection limit of 25 IU/L.  That is, a sample with an hCG concentration above 25 IU/L should produce a positive result.  Of the samples that gave a positive result, about 20% had an hCG concentration that was <25 IU/L which indicated that the qualitative test was more analytically sensitive than we expected it to be.  In my opinion, that’s a good thing.

Because we determined the actual pregnancy status of all the patients with a positive result, we were also able to determine how well the qualitative and quantitative tests performed at determining pregnancy status.  Both tests did quite well and showed high sensitivity and specificity.  That is, there were very few false-negative or false-positive results.  From a clinical perspective, a false-negative result is more concerning than a false-positive one because a pregnant patient who is incorrectly identified as not being pregnant risks being exposed to a medical intervention that could harm the fetus.  The false-negative rate was lowest, only 0.1%, when the qualitative test was evaluated against pregnancy status and the detection threshold of 25 IU/L.  The performance of the quantitative serum hCG test was identical.  So, both the qualitative and quantitative serum hCG tests do a very good job at ruling-out a possible pregnancy.

So, given this evidence, I would conclude that while qualitative hCG tests could be replaced by quantitative tests, there is really no compelling reason to do so.

Four reasons for a positive hCG test in the absence of pregnancy

“Why is the pregnancy test positive if she’s not pregnant?”

This is a question I’ve been asked several times and it’s a good one.  The query usually comes from a nurse, doctor, or other healthcare provider after performing a test for human chorionic gonadotropin (hCG) and getting a positive or elevated result that they did not anticipate.

Because hCG is a hormone normally produced during pregnancy, hCG tests are used to diagnose the pregnant patient.  That makes it easy to think of hCG tests as “pregnancy tests.”  While that’s not an inaccurate label for them, identifying hCG tests as pregnancy tests gives the impression that is all they are supposed to do.  Technically, hCG tests are designed to qualitatively detect and/or measure the hormone in urine or blood and there are other reasons besides being pregnant that can cause hCG to be present.

I can think of four different reasons why hCG could be present in a non-pregnant woman.

1. Biochemical pregnancy.  A biochemical pregnancy occurs when a woman becomes pregnant yet has a spontaneous loss of the fetus before she even knew she was pregnant.  If hCG testing occurs before all of the hCG has been metabolized out of the body then hCG can be detected by a lab test.  This situation is not as uncommon as one might think for two reasons.  First, hCG tests are frequently performed in healthcare settings in order to identify the pregnant patient in order to avoid any medical interventions that are potentially harmful to a fetus.  Second, hCG tests are capable of detecting very low concentrations of the hormone.  The high frequency of testing combined with the analytical sensitivity of the tests means that biochemical pregnancies are easily detected.
2. Pituitary hCG.  Although the placenta normally produces hCG during pregnancy, it can be made by the pituitary gland.  The pituitary gland is a small structure in the brain that secretes many different hormones that function to regulate many endocrine organ systems.  Interestingly, three hormones normally produced by the pituitary gland (thyroid stimulating hormone, follicle stimulating hormone, and luteinizing hormone) are structurally similar to hCG.  Pituitary hCG is more commonly detected in women greater than 55 years of age but can be detected in women as young as 41 years.  Non-pregnant women with pituitary hCG usually have low concentrations of hCG present in the blood and urine.  Importantly, concentrations of hCG produced by the pituitary gland don’t show the rapid increases that occur during pregnancy.
3. Malignancy.  Cancer cells sometimes make hCG.  While many different types of cancer have been shown to make the hormone, it’s most commonly associated with the gestational trophoblastic diseases and certain types of germ cell tumors of the testes.  Because testicular tumors occur only in men, the question of detecting hCG in the absence of pregnancy is clearly not relevant.
4. Interfering antibodies.  Some women have antibodies in their blood that can interfere with hCG tests and cause a positive or elevated result in the absence of hCG.  Only hCG tests performed on blood can be affected by this problem because the interfering antibody molecules aren’t normally present in the urine.  This can be a serious problem because some women have been mistakenly diagnosed with cancer due to the false-positive hCG test result and have undergone unnecessary treatments for it.  The frequency of this problem is difficult to know but it’s probably very low.  Over the last several years, the manufacturers of hCG tests have worked to minimize possible interference from these antibodies but nothing can be done to completely eliminate the problem.  When alerted, the laboratory can help to determine if an hCG test result is falsely positive due to this issue.

So, just because an hCG test result is interpreted as positive doesn’t automatically mean that a woman is pregnant.  There are very valid reasons for detecting hCG in the absence of pregnancy.  That said, when the hCG test result doesn’t match the clinical picture, the laboratory should still be asked that question!  When alerted to the discrepancy, the lab can help to investigate the problem and perhaps shed some light on the cause.

There is quite a bit more to say on each of those four causes but I’ll save those comments for future posts.

 

Wanted: a rapid pregnancy test using whole blood

Most laboratorians would agree that when it comes to hCG testing for pregnancy status, a blood sample is preferred over a urine sample.  Why is that?

• Following implantation of a fertilized egg, hCG is first detected in the blood.  hCG in the urine usually appears a day or two later.
• Blood-based hCG test results provide a concentration of the hormone whereas urine tests are qualitative (positive or negative)
• In early pregnancy, when hCG concentrations are still relatively low, a negative result can be obtained from a dilute urine sample (i.e. a false-negative result).

An earlier post presented the pros and cons of the different types of pregnancy tests. To summarize here: qualitative (yes/no) tests are typically easier to perform and, if urine is the sample being tested, then qualitative tests can truly be performed at the point-of-care.  If a blood sample is used, it has to be centrifuged to separate the liquid part of the blood (serum) from the solid part (blood cells), a step that is nearly always performed in the laboratory.  As such, qualitative hCG tests using a blood sample cannot be performed at the point-of-care.

A recent study challenges that idea.

Blood samples were collected from 632 women (aged 18-51 years) presenting to an emergency room. These blood sample were tested for hCG in two ways: 1) traditional, quantitative serum hCG testing in the laboratory and 2) qualitative, point-of-care hCG testing in the emergency room using the whole blood sample instead of serum.  Note that the qualitative test used was not designed nor approved by the FDA for use with whole blood and that the test results were interpreted after 10 minutes whereas the test instructions required result interpretation after 5 minutes.

When compared to the results of the quantitative test performed in the lab, the qualitative test performed on whole blood (instead of serum) performed quite well.  100% of the laboratory-negative results were interpreted as negative using the qualitative test and 96% of the laboratory-positive results were interpreted as positive using the qualitative test.  The hCG concentration of the 4% of "false-negative" results (9 out of 216) ranged from 5 to 50 IU/L.  Notably, the hCG detection threshold of the qualitative test is identified as 25 IU/L.

So, what does all this mean?  Well, the authors of the study suggest that it can be difficult to obtain urine samples from some patients which makes testing hCG in blood more attractive.  But because blood-based hCG tests are normally always performed in the laboratory, waiting for the lab results can waste time.  Hence, the appeal of testing whole blood for hCG at the point-of-care.  The results of this limited study suggest that, perhaps, whole blood hCG testing using devices that were designed for urine or serum, has merit.

What the study did not address, however, are the (often) complicated state and federal rules that regulate lab testing.  The study's use of a qualitative hCG test using whole blood instead of serum is problematic.  Changing the type of sample recommended by the test manufacturer is a thorny issue for laboratorians who are ultimately responsible for all lab tests performed.  It is typically the lab's responsibility to determine if an alternate sample type can be used for a given laboratory test and this type of substitution, however appealing, cannot be widely adopted without a thorough evaluation and validation by the laboratory.

Intriguing?  Yes.  Easily implemented?  Not quite.

Surprising details about the most sensitive urine pregnancy tests

Here’s a multiple-choice question to consider:

Which type of qualitative urine hCG test devices (a.k.a pregnancy tests) are capable of detecting the lowest concentrations of hCG?

A. The devices that can be purchased at a local pharmacy for use at home (over-the-counter (OTC) tests).

B. The devices used in hospitals, clinics, and doctor’s offices (point-of-care (POC) tests).

C. Both types of devices are equivalent.

The correct answer is A.

Surprised? I was too. Logic suggests that the urine hCG tests that are frequently used in healthcare settings (versus home use) would probably do best at detecting the hormone at lower concentrations.

There have been two studies that have investigated the analytical sensitivities of urine hCG tests. One study used samples prepared by adding hCG to male urine to achieve hCG concentrations of 6.3, 13, 25, and 50 IU/L. Fifteen different OTC pregnancy tests were evaluated, twelve of which claimed an hCG detection limit of 50-100 IU/L (the other three claimed 25 or 40 IU/L). All but one of the fifteen devices produced positive results at an hCG concentration of 25 IU/L. Curiously, the one device that did not claimed a detection limit of 25 IU/L.

The other study (disclaimer: I was one of the investigators) used urine samples obtained from pregnant women within 10 days of their expected period. In my opinion, this is the optimal type of sample to use because pregnancy urine contains a mixture of different hCG variants and not just a single form of the hormone. That is important because not all pregnancy tests detect the same variants of the hormone equally and that, in turn, can influence device sensitivity. These urine samples were diluted to low hCG concentrations and then tested with both OTC and POC pregnancy tests. While all the devices claimed an hCG detection limit of 25 IU/L, we observed that they were capable of detecting hCG at much lower concentrations. The POC devices were sometimes four times more sensitive than claimed and the OTC devices were sometimes sixteen times more sensitive!

The greater analytical sensitivity of the pregnancy test devices appeared to be a function of the urine sample used. For example, using the hCG Combo device, urine samples from five different women produced positive results at 25 IU/L, while samples from two women produced positive results at a lower concentration (6 and 13 IU/L) and samples from three women produced positive results at a higher concentration (50 IU/L). Likewise, the OTC devices also showed variability in analytical sensitivity with urine samples obtained from different women although all of them produced positive results at hCG concentrations below the claimed 25 IU/L detection limit. In other words, the OTC pregnancy tests gave positive results at lower hCG concentrations than did the POC pregnancy tests!

We suspect that home pregnancy tests have greater analytical sensitivity than their POC counterparts because of differences in the amount of urine the devices require. OTC devices absorb urine from the urine stream during voiding and, as such, they use a much larger volume of urine than the tests used in healthcare settings (approximately 5-10 times more).

To be certain, the physician who orders a urine hCG test on a patient has the same desire as the woman who performs her own urine test at home: they both want to know about pregnancy status. In the former scenario, the test is often being done to rule-out a possible pregnancy prior to a medical intervention that could harm a fetus. In the latter, the test is being done to satisfy a curiosity. Which test do you think should be the most sensitive?

An antiquated pregnancy test?

Many people are surprised at how many different types of "pregnancy tests" are available.  The use of quotations is deliberate because pregnancy tests don't detect pregnancy, they detect a hormone produced during pregnancy: human chorionic gonadotropin (hCG).  But, because hCG is usually only present during pregnancy, one can forgive the non-specific label.

Here's a table that illustrates the available hCG tests:

So hCG tests can be performed on urine or blood using a qualitative (yes/no) or a quantitative (concentration) test.  Because there is no clinical need to quantify hCG in urine, quantitative tests are usually not performed in urine.  Some laboratorians (yours truly included) do believe that quantitative urine hCG tests can be helpful in specific situations but that's a post for a different day.

In this post, I want to focus on the qualitative hCG test performed with a blood sample.  A recent paper suggested that this test should be considered antiquated (disclaimer: I was an author on that paper).  There are a few reasons why this might be true:

1. The biggest value of qualitative hCG tests is that they can be performed at the point-of-care with test results available to the physician very quickly.  However, blood samples have to be centrifuged to separate the liquid part of the blood (serum) from the solid part (blood cells), and the centrifugation step is nearly always performed in the laboratory.  As such, qualitative hCG tests using a blood sample cannot be performed at the point-of-care.
2. The vast majority of clinical laboratories can perform quantitative serum hCG tests.  Although these tests may take a bit longer to perform than qualitative tests, the time it takes to perform them is still rather fast (~20 minutes).
3. Quantitative hCG tests can probably measure hCG at a concentration lower than the detection thresholds claimed by qualitative tests (~1 IU/L vs. 10-25 IU/L, respectively).  If that's the case, why not use the most sensitive hCG test for a blood sample?
4. The only clinical use of a qualitative serum hCG test is to detect (or rule out) a possible pregnancy.  Therefore, it can be argued that the most analytically sensitive test available (i.e a quantitative test) should be utilized.

Given those reasons, I've been wondering if qualitative blood tests for hCG could be abandoned.  Why bother using them if quantitative tests are readily available and can detect hCG at lower concentrations than qualitative tests?  Seems like a no-brainer to me but, as a scientist, I need data to support my hypothesis.  This is something to investigate!

False Negative Pregnancy Tests

A false negative pregnancy test means that the test tells the patient they are not pregnant when they really are. Why does this occur? There are four well known reasons pregnancy tests can give false negative results.

1) The most common reason is testing too early after fertilization occurs. The pregnancy hormone, hCG, is not produced until implantation occurs and it takes several days for the hCG concentrations to get high enough in blood and then urine to give a positive signal. Most devices give positive results around the day of a woman's expected period, but this can vary widely.

2) Another reason for false negative results is dilute urine. If a woman drinks a lot of fluids, the concentration of hCG in the urine will be more dilute. For this reason many doctors recommend that testing be performed on the first urine of the morning because this tends to be the most concentrated (because you probably have not drank anything all night long).

3) A very rare cause of false negative results occurs when very, very, high concentrations of hCG are present. This is called the high-dose hook effect. The hCG assays works by forming a so-called "sandwich" with two different antibodies as the "bread" and the hCG molecule as the "meat." The hook effect occurs when the hCG concentration is so high that it saturates both antibodies and there are so many molecules that the antibodies don't actually form a sandwich. This is rare because women don't normally produce enough hCG to saturate both antibodies. The hook effect should be of concern in a hospital setting, but most women should not be concerned about a hook effect with their urine. A hook effect can be confirmed if testing shows a positive result after sample dilution.

4) Finally, the other reason for false negatives was only recently described and is referred to as the "variant hook effect." This is much more common than the hook effect. As pregnancy progresses, there are actually different variant forms of hCG that begin to appear in the urine. After about 5 weeks of pregnancy (i.e. 3 weeks after the expected period) concentrations of hCG beta core fragment are higher than all other forms of hCG. This is perfectly normal. Unfortunately, the concentration of hCG beta core fragment can saturate one of the antibodies used in the assay in certain pregnancy kits, and the other antibody doesn't recognize the beta core fragment. As a result, no sandwich forms and the test is read as a negative. The farther in pregnancy a woman is, the more likely that this false negative will occur. Similar to the hook effect, the variant hook effect can be confirmed if testing shows a positive result after diluting the sample.

Posted by Ann M. Gronowski, PhD 5/8/11

False-positive results from qualitative serum pregnancy tests

OK.  I know I've written a lot about false-positive hCG tests already, both in February and earlier this month. Bear with me as I still have more to say on this issue.

I spoke on the topic of hCG testing at a clinical lab science conference a couple of weeks ago and got this question from someone in the audience:

"I was told that one way to investigate a possible false-positive, quantitative hCG result obtained from a serum sample was to test the sample using a qualitative hCG test because interfering antibodies don't affect those tests; is that true?"

(If you are unfamiliar with the concept of interfering antibodies and want to learn more, read this.)

My response to this question was "No, that is not true."  I'm uncertain why one would think that qualitative serum hCG tests were somehow immune to any influence from interfering antibodies but that wasn't the first time I had fielded that same question.  Clearly, this is not an isolated misconception.

To support my response that qualitative tests can also produce erroneous results due to interfering antibodies I referred to a case report we described a short while ago.  You can find that report here.  The case describes a 46-year-old woman who was not sexually active yet whose blood sample produced a positive result using a qualitative serum pregnancy test.  Because the positive result was unexpected, her doctor asked us to measure the concentration of hCG in the same serum sample.  That result was less than 2 IU/L (normal is less than or equal to 5).

When we repeated the qualitative test after treating the serum sample with a blocking agent that removes interfering antibodies, the result was interpreted as negative.  A urine sample obtained from the patient also produced a negative qualitative hCG test result.  The logical conclusion was that an interfering antibody produced a false-positive result with the qualitative test.

I don't know how often this scenario might occur but it's clear that labs shouldn't rely on the notion that a qualitative hCG test is immune to any influence from interfering antibodies.

Investigating unexpectedly positive hCG test results

In February, I wrote about interfering antibodies being the cause of false-positive hCG results from blood samples.  As a follow-up, I thought it would be a good idea to talk about what the lab can do to investigate the potential problem.

As a reminder: false-positive hCG results are often attributed to the presence of an interfering antibody in the blood sample.  Such was the case with Jennifer Rufer who was misdiagnosed with cancer due to a false-positive hCG blood test.  It's important to know that laboratories are likely not able to independently identify when an interfering antibody is present.  An interfering antibody should be suspected whenever the clinical picture of the patient fails to match the laboratory (in this case, hCG) results.  When laboratorians are asked to investigate the possibility of a false-positive hCG result there are several ways to do so but the key piece of information to note is that the lab has to be notified that there is the suspicion of an erroneous hCG result.  This is because the laboratory is nearly always unaware of the clinical picture of the patient.  We need to rely on our clinical colleagues to alert us to the possibility that an hCG test result is potentially incorrect.

So, what can the lab do when asked if an hCG blood test result is falsely increased?  There are several approaches the lab can take to investigate:

1. Perform an hCG test on a urine sample obtained from the patient.  Because hCG is excreted in the urine, the detection of hCG in a urine sample indicates that the result from the blood sample is likely accurate.  Interfering antibodies aren't excreted into the urine and so if hCG is detected in the urine then the blood test result is probably really due to the presence of hCG.
2. Perform a dilution of the blood sample.  Because interfering antibodies are reactive against the hCG assay reagents, the expected response to sample dilution is typically not observed.  That is, if the sample is diluted by a factor of 1:2, the hCG concentration should decrease by a factor of 2 if the hCG molecule is truly present.  Failure to observe the expected decrease in concentration supports the presence of an interfering antibody.
3. Repeat the hCG test using a different method.  Interfering antibodies may be reactive against antibodies derived from a specific animal species (e.g. against mouse antibodies).  If the hCG test is repeated using a different method and the results are considerably different then that can suggest the presence of an interfering antibody.  Importantly, the alternative method selected should be one that uses antibodies derived from a different species of animal than the test in question.
4. Treat the sample with blocking agents.  Blocking agents are commercially available that can be used to remove (by adsorption) potentially interfering antibodies from the sample.  If the hCG results after treatment is considerably different from the result of the untreated sample, then interfering antibodies may be present.  The definition of considerably different is not well defined although many labs use a difference of 50% to alert them to the possibility that interfering antibodies are present.

I recommend that laboratories use more than one of these investigations rather than rely on a single one.  As the Rufer case so clearly demonstrates, erroneous laboratory results can result in serious harm to patients.  Together with our clinical colleagues, it our responsibility to do all that we can to assure that laboratory test results can be correctly interpreted.

False positive blood pregnancy test results

Today I gave a lecture on hCG testing to 4th year medical students at the University of Utah.  Part of that lecture included the tragic ordeal of Jennifer Rufer.  In 2001, she was awarded nearly 16 million dollars because of a misdiagnosis of cancer from a false-positive pregnancy test.  I previously wrote about false-positive pregnancy tests from urine specimens.  This post deals with false-positive results from blood tests.

The same things that can cause false-positive results from urine specimens can also cause false-positive results from blood samples.  It might help to review those first.  There is another cause of false-positive hCG test results that is unique to blood samples: interfering antibodies.

First, a quick summary:

• Human blood normally contains many different types of antibodies: molecules made by our immune system that identify and neutralize invaders such as bacteria and viruses.
• Animals make antibodies, too, and their antibodies are used to make the components of hCG tests.  For example, mouse antibodies that recognize human hCG molecules are used in tests that measure the concentration of hCG.  One mouse antibody binds to one part of the hCG molecule (capture antibody) and a second mouse antibody binds to a different part of the hCG molecule.  That second mouse antibody has a signaling molecule attached to it.  We can measure the strength of that signal and that signal strength will be equal to the amount of hCG that it is bound to:

• Some people's immune systems make antibodies that react against animal antibodies.  These "anti-animal" antibodies can cause false-positive hCG test results because they connect the capture antibody and the signaling antibody together even when hCG is not present, like this:

This was the cause of Jennifer Rufer's false-positive pregnancy test.  Unfortunately for her, the false-positive result was recognized for what it was only after she had been treated for what her doctors thought was an aggressive type of cancer called choriocarcinoma.

It's uncertain how many people have interfering antibodies in their blood.  Some estimate the frequency as high as 10-20%, other studies put that number much lower at <1%.  It doesn't really matter how many people have them but what does matter is how doctors and laboratorians work together to identify this problem when it occurs.  That's easier said than done.

Why?  Because the laboratory that does the testing is nearly always unaware of the medical history of the patient so it's practically impossible for the lab to know if an hCG result is inconsistent with the patient's condition.  The lab has to rely on the doctor to question an hCG test result when it don't fit what they are seeing in their patient.  If the hCG result doesn't fit the clinical picture and the doctor calls the lab for help, there are investigations the lab can do to help identify if an interfering antibody is present or not.  But without that call questioning the result, the lab doesn't have reliable mechanisms for identifying all cases of interfering antibodies.

 

 

False positive urine pregnancy test results

In an earlier post, I wrote about the different types of tests used to detect hCG. While the results of These test are usually very accurate, false positive results can and do occur.  Here I discuss some reasons why a urine pregnancy test might give a false positive result.

First, though, a few comments about what a false positive result actually is.  Pregnancy tests are designed to detect hCG;the hormone produced by the placenta during pregnancy.  If the hormone is truly present in the urine and the test detects it, then that result is a true positive result.  In other words: hCG present = true positive test result.  However, because hCG is usually only produced during pregnancy,many people consider a positive result to mean that a woman is pregnant.  There's a big difference between those two meanings!  To summarize:

• Pregnancy tests detect hCG
• hCG is usually (but not always) produced during pregnancy
• A positive pregnancy test indicates that hCG is present (even if the hCG is being made by something other than the placenta)

With that in mind, let's explore the causes of false positive urine pregnancy test results.

1. Missed reaction time:  The result of a pregnancy tests should be read when the manufacturer instructions say to read it.  If the result is read after that time, a faint line is sometimes visible (possibly due to evaporation of the urine) that might be interpreted as a positive result.   This would truly be a false positive result because hCG is not present.
2. Biochemical pregnancy:  A biochemical pregnancy is a real pregnancy but is one that ends very early after conception.  If a pregnancy test is performed near the time of the expected period (or even several days later), a positive result can be obtained.  This is actually not a false positive result because hCG is present and it was detected by a test designed to detect it!  Even though the pregnancy has ended, hCG can still be detected in the urine for a few, or even several, more days.  Biochemical pregnancies are quite common although many aren't even detected because the miscarriage occurs before the woman even knew she was pregnant and so had no reason to perform a pregnancy test.  Indeed, the term "biochemical pregnancy" came into use only after urine pregnancy tests became sensitive enough to detect hCG close to the day of the expected period.  Prior to that, the tests were not able to detect that rather low amounts of hCG that are associated with a biochemical pregnancy.
3. hCG from sources other than the placenta:  Although hCG is most commonly made only during pregnancy, there are other times that it is produced.  Certain cancers sometimes make hCG; most notably a family of tumors collectively called trophoblastic tumors.  The pituitary gland in the brain can also produce hCG although this is more common in women who have gone through menopause.  Finally, people taking hCG as a weight-loss aid (a waste of money, by the way) or as part of fertility treatments can also have detectable hCG in their urine.