| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
Department of Pathology, University of Utah School of Medicine, and ARUP Laboratories, Inc., Salt Lake City, UT 84108.
a Address correspondence to: ARUP Laboratories, Inc., 500 Chipeta Way, Salt Lake City, UT 84108. Fax 801-584-5207; e-mail ashwood{at}med.utah.edu
 |
Abstract |
|---|
 Top Abstract IntroductionPreanalytical ConsiderationsAnalytical ConsiderationsPostanalytical ConsiderationsReferences |
|---|
Key Words: indexing terms: guidelines • fetal organ maturity • amniotic fluid chemistry • phosphatidylglycerol • lecithin/sphingomyelin ratio • fluorescence polarization
|
Introduction |
|---|
|
TopAbstractIntroductionPreanalytical ConsiderationsAnalytical ConsiderationsPostanalytical ConsiderationsReferences |
|---|
|
Preanalytical Considerations |
|---|
|
TopAbstractIntroductionPreanalytical ConsiderationsAnalytical ConsiderationsPostanalytical ConsiderationsReferences |
|---|
During amniocentesis, the location of the needle will affect the result of the FLM test. Pulmonary surfactant concentration will be higher in the fluid surrounding the fetal mouth. Thus, samples obtained close to the fetal head will be more mature than those obtained at other sites. Because this effect is subtle, I recommend that FLM results not be interpreted with respect to location of specimen acquisition.
Samples obtained from the vaginal posterior fornix after rupture of the membranes are commonly contaminated with blood, bacteria, and mucus. Vaginal pool samples are adequate for testing only when the fluid has been in the vagina a short time and the sample is quickly chilled after collection (14). I recommend that the clinician seriously consider amniocentesis in patients with ruptured membranes. Often, a pocket of amniotic fluid can be located by ultrasound and collected by amniocentesis without endangering the fetus. This type of sample is much preferred over a vaginal pool sample.
Amniocentesis trays generally contain an amber collection tube. While this amber tube is satisfactory for the collection of FLM specimens, it is not required. The phospholipids present in amniotic fluid are not light sensitive. The amber tube is required if the measurement of amniotic fluid bilirubin is anticipated, as in a case of erythroblastosis fetalis.
After collection, I recommend that the specimen tube be immersed in wet ice for transportation to the laboratory. Chilling the sample reduces the activity of any enzymes that might be present and could degrade the phospholipids.
sample handling
Most samples for FLM assessment are tested the day of collection.
When testing must be delayed, refrigeration at 4 °C is appropriate.
Before testing, samples should be mixed well, assessed for
contamination, and, if required by the method, centrifuged. For
long-term studies, centrifuged amniotic fluid can be stored at
-20 °C or -70 °C.
Surfactant phospholipids are present in amniotic fluid in fairly large particles, lamellar bodies. These bodies are ~0.5–1.0 µm in diameter (15). Their density is greater than that of the aqueous phase of amniotic fluid. Therefore, the lamellar bodies will settle to the bottom of the collection container if the specimen is allowed to stand undisturbed. Samples must be thoroughly mixed before splitting, aliquoting, or testing. Vortex-mixing should be avoided because it causes the samples to foam, especially when the samples are mature. I recommend that amniotic fluid samples be mixed by placing them on a hematology tube rocker for not less than 2 min. The sample should not be allowed to stand after mixing, but instead immediately processed.
A note should be made if the sample contains blood, meconium, bilirubin, or mucus. This can be best assessed if the entire sample is decanted from the usually amber collection tube into a new clear specimen tube. The results of most methods will be affected by a 1% or greater contamination with blood. The best way to visually estimate this amount of contamination is to add 10 µL of EDTA–whole blood to a 1-mL amniotic fluid specimen for comparison. Green-tinged fluid indicates the presence of meconium or old blood. If heavily contaminated, the specimen will contain a large amount of particulate matter. Specimens with a yellow tint indicate the presence of bilirubin. Many vaginal pool specimens will contain mucus, which can be assessed while decanting from the collection tube into the new specimen tube.
If the analytical method requires a centrifugation step, then great care should be taken to follow the protocol exactly. My laboratory uses 400g for 2 min. This will remove red blood cells, large clumps of mucus, and, unfortunately, ~8% of the lamellar bodies. Always recalculate the centrifugal speed needed to achieve the required centrifugal force when using a new centrifuge or a different rotor. Longer centrifugation periods and higher forces remove more surfactant.
|
Analytical Considerations |
|---|
|
TopAbstractIntroductionPreanalytical ConsiderationsAnalytical ConsiderationsPostanalytical ConsiderationsReferences |
|---|
15 L/S ratio requests per
week, I recommend either performing the test in-house or
referring it to a reference laboratory. The Helena method for L/S ratio
testing is a satisfactory one-dimensional method. All laboratories that support obstetric patients should offer at least one test that can be performed in <1 h. The choices are: the FP method on the TDx (either with the Abbott TDx/FLM® test, which reports in milligram of surfactant per gram of albumin, or the home-brew version that involves fluorescently labeled phosphatidylcholine and has units of polarization), PG with the slide method (AmnioStat-FLM® from Irvine Scientific), or the foam stability test (either with the Beckman Lumadex®-FSI, or a home-brew version that involves ethanol and amniotic fluid mixtures at differing concentrations).
The FP method on the TDx is the best rapid test in a high-risk setting. This test can be easily offered 24 h per day, every day of the week. The results are quantitative and changes in the results indicate changes in the risk of RDS. Precision is good as shown by a recent CAP proficiency survey (sample identification, mean, SD, and CV): LM-11, 114.7 mg/g, 5.7 mg/g, 5.0%; and LM-12, 19.9 mg/g, 0.7 mg/g, 3.9% (16). The Abbott and home-brew FP methods produce equivalent results (6).
The Irvine Scientific slide test for PG works best in a low-risk population because of the poor predictive value of an immature result. A disadvantage of this test is the qualitative nature of the result: negative, weakly positive, or positive. Thus, with a negative (immature) result, a clinician will not be able to determine if the fetal lungs have started to mature. A major advantage is that PG is unaffected by blood contamination. Accuracy is good as assessed by a recent CAP proficiency survey (sample identification and percent correctly classified): LM-09, 98.6% positive; and LM-10, 93.9% negative (16).
Finally, laboratories should consider offering LBC (9)(10)(11). This test takes <15 min and is performed by counting particles in the platelet channel of a hematology blood analyzer. Most studies have used a Coulter Counter. The results may differ on other brands of blood cell counters because of size interval differences in the platelet counting channel. Because of the small number of prospective clinical outcome studies reported, the results of this test should be reported in parallel with one of the rapid tests above.
frequency of testing
Laboratories performing L/S ratio should offer routine testing
once per day, 7 days per week. Emergency requests for this test should
be discussed with the laboratory director or supervising technologist.
All other FLM tests should be performed daily in a routine run and be
available at any time for emergency testing without prior approval. The
turnaround time for these rapid tests should be <1 h.
|
Postanalytical Considerations |
|---|
|
TopAbstractIntroductionPreanalytical ConsiderationsAnalytical ConsiderationsPostanalytical ConsiderationsReferences |
|---|
interpretation of results
L/S methods can vary substantially. Therefore, laboratories
performing L/S ratio testing should either conduct a clinical outcome
study to validate their reference intervals or compare their method
with one in the literature that does have a clinical outcome study. The
FP test on the TDx, the slide-based PG, and the FSI are better
standardized than the L/S ratio. Laboratories should use with
confidence the literature-based reference information for these rapid
tests.
Care should be taken to avoid misrepresenting the predictive value of a mature or immature test. Most recent clinical outcome studies appropriately evaluate the FLM test results by using a ROC curve (17). This is a plot of the sensitivity (percent of affected cases having an immature result) vs 1 - specificity (percent of unaffected cases having an immature result). The predictive value of any given result is strongly affected by the prevalence of respiratory distress for that healthcare facility. The prevalence can vary from 3% to 15% depending on the acuity of the patients. Thus, statements regarding the predictive value of a mature test may be accurate for one healthcare setting and quite biased in another. Logistical models to predict the risk of RDS have been proposed that include the FLM result and the gestational age (18)(19). Theoretically, these models include gestational age to represent both the prior probability of RDS and other factors that affect the FLM result. If these models are confirmed by other investigators, they may be useful for interpreting FLM results across differing healthcare settings.
The dogma in FLM testing teaches that diabetes delays pulmonary maturation and makes the FLM test results unreliable. This belief comes from older studies that showed immature FLM test results in term diabetic pregnancies. Pregnancy management for patients with diabetes has, in the past decade, emphasized tight glycemic control. Poorly controlled diabetes clearly delays pulmonary maturation (20). Well-controlled diabetes very likely does not delay maturation nor make the FLM test results unreliable. There is no clinical outcome study in the literature with enough RDS cases from diabetic mothers to prove that the tests are unreliable. There is indirect evidence that the FP test is reliable (5)(19)(21). Thus, I recommend against using separate reference intervals for diabetic patients.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionPreanalytical ConsiderationsAnalytical ConsiderationsPostanalytical ConsiderationsReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  C. R. Fantz, C. Powell, B. Karon, C. A. Parvin, K. Hankins, M. Dayal, Y. Sadovsky, V. Johari, F. S. Apple, and A. M. Gronowski Assessment of the Diagnostic Accuracy of the TDx-FLM II to Predict Fetal Lung Maturity Clin. Chem., May 1, 2002; 48(5): 761 - 765. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
