College of American Pathologists
Printable Version

  Q & A






January 2006

Q: Why are there so many published equations for calculating osmolality and is one equation best? Should the laboratory provide a calculated osmolality in addition to the measured value?

A. Osmolality is a colligative property of solutions that is dependent on the number of dissolved particles in solution. The concentration of solute causes a change in certain physical properties of the solution: the freezing point and vapor pressure are decreased and the osmotic pressure and boiling point are increased. In dilute solutions, there is a linear change in these colligative properties as the solute concentration increases.

Serum osmolality is most often assessed by measuring the decrease in freezing point that results from the presence of solute. The addition of 1 mole of a non-dissociating solute to 1 kilogram of water will depress the freezing point by 1.858°C. Another, less frequently used, technique for determining osmolality is to measure the depression in vapor pressure. The greater the number of dissolved particles present, the lower the vapor pressure of the aqueous component of the solution. An important exception is when the solute itself is a volatile substance, such as ethanol. Although osmolality should be expressed as milliosmoles per kilogram (mOsm/kg) of water, it is usually reported as milliosmoles per liter (mOsm/L) of water. This difference in reporting introduces a slight, clinically insignificant error.1

Three types of solutes (osmolutes) are most often encountered in biological fluids: electrolytes, organic non-electrolyte molecules, and colloids.2Serum osmolality is determined primarily by five major osmolutes: sodium, chloride, bicarbonate, glucose, and urea. In serum, sodium comprises the majority of the cations present. Since each sodium ion can be assumed to be counterbalanced by an anion, such as chloride and bicarbonate, only glucose and urea need to be measured, in addition to sodium, to calculate the osmolality.

A stat procedure frequently used to estimate the osmolality of serum is the calculated osmolality. The calculated osmolality is not meant to supplant the measured osmolality but instead should be used in acute emergency situations only. Some equations that have been proposed for calculating osmolality were developed for use in the general patient population, while others were developed using specific patient populations, including those with diabetic ketoacidosis, hyperosmolar non-ketotic coma, hemorrhagic shock, and postsurgery.2

The level of agreement between measured and calculated osmolality depends more on the population tested than the equation used. Patients who have increased concentrations of unmeasured solutes, such as those with renal failure, will show greater differences between measured and calculated osmolalities. This must be taken into consideration when interpreting a calculated osmolality. The calculated osmolality also will give falsely low results when volatile solutes are present. The difference between osmolality measured by a freezing-point depression osmometer and calculated serum osmolality has been suggested as a rapid test for the presence of ingested volatiles, especially alcohols.3

At least 16 equations to calculate serum osmolality have been developed using linear regression analysis or other statistical methods.1,3,4 Proposed equations include those as simple as multiplying the sodium by some factor,5 to equations that incorporate sodium, potassium, calcium, magnesium, glucose, and urea.4 Including these additional analytes does not result in a more accurate calculated osmolality. Three commonly used equations for calculating osmolality are noted above. When glucose and BUN are reported in terms of mass concentrations (mg/L), these analytes must be divided by their formula equivalent weights, 180 and 28, respectively, to convert their mass concentrations to millimoles per liter. While all three equations correlate extremely well with measured osmolality (r > 0.96), equation2 has been shown to have the smallest bias from the measured osmolality.1 Equation2 is designed for use with analyzers or systems that can provide automated calculations, while equation3 allows for ease of calculation.


  1. Dorwart WV, Chalmers L. Comparison of methods for calculating serum osmolality: from chemical concentrations, and prog nostic value of such calculation. Clin Chem. 1975;21:190-194.
  2. Gennari FJ. Serum osmolality, uses, and limitations. N Engl J Med. 1984;310:102- 105.
  3. Bhagat CI, Garcia-Webb P, Fletcher E, et al. Calculated versus measured plasma osmolality revisited. Clin Chem. 1984;30:1703-1705.
  4. Rasouli M, Kalantari KR. Comparison of methods for calculating serum osmolality: multivariate linear regression analysis. Clin Chem Lab Med. 2005;43:635-640.
  5. Winters RW. Disorders of electrolyte and acid-base metabolism. In: Barnett L, ed. Pediatrics. 14th ed. New York, NY: Appleton-Century-Crofts; 1968:336-338.

Steven Kazmierczak, PhD, DABCC
Professor of Pathology
Department of Pathology
Oregon Health and Science University

Liaison, CAP Chemistry
Resource Committee

Q: At one of the 25 hospitals in our company, the pa thol ogist reviews more than 70 percent of the hematology slides and makes a report. This process slows down the turnaround time for reporting hematology differential results. Is this practice common and acceptable? Pathologists at the 24 other hospitals do very few peripheral smear reviews with reports.

A. It is not appropriate to comment on a specific practice pattern without considerably more detail about the setting and the types of patients being evaluated. Therefore, I will offer general comments about the need to perform differential slide reviews and when a physician should be involved in these slide reviews. Perhaps these guidelines might serve as a basis for your pathologists to develop a consistent approach to differential slide review.

When a hemogram with differential is requested, it is generally agreed that if all the parameters are within laboratory-defined acceptable ranges and there are no instrument-generated flags on the instrument (automated) differential, then the most cost-effective and efficient approach is to provide a report of the parameters and instrument differential without anyone reviewing the smear. A laboratorian with the appropriate training and expertise should review a blood smear if the instrument yields a parameter outside laboratory-defined acceptable ranges, there is an instrument flag, or a physician specifically requests that a smear be reviewed—for example, to evaluate a smear for evidence of malaria.

Some laboratories always perform and report a manual differential smear result in these circumstances—that is, the smear is reviewed by a physician only if such review is indicated by laboratory policies. Other laboratories perform a blood smear scan to confirm that the instrument differential appears correct and that there are not significant abnormalities or issues with the instrument-generated report and, if so, report the parameters and instrument differential.1 This use of the manual smear review as a validation procedure rather than an automatic substitute for automated methods again is viewed as a more cost-effective and efficient approach to patient care.

Laboratories must set criteria for when a physician needs to be involved in the review and final reporting process. The CAP Hematology/Clinical Microscopy Resource Committee offers a list of indicators for when physicians should review blood smears.2 They recommend such a review based on first observation of the following on a blood smear:

  • abnormal cells
  • abnormal red cell morphology
  • clinical or other suspicion of organisms manifest in a blood smear
  • dysplasia in any cell lineage
  • inclusions in red cells or platelets
  • intranuclear or cytoplasmic inclusions in leukocytes
  • pancytopenia or other marked cytopenias

The committee further recommends that blood smears be reviewed based on subsequent observations of the following:

  • persistence or recurrence of abnormal cells
  • persistence of organisms

When to generate a specific consultative report, as opposed to simply reporting the observed abnormality within the hemogram and differential report, should be based on a consistent laboratory policy, usually developed in concert with the lab’s clinicians.


  1. Lantis KL, Harris RJ, Davis G, et al. Elimination of instrument-drive reflex manual differential leukocyte counts. Am J Clin Pathol. 2003;119:656–662.
  2. Peterson P, Blomberg DJ, Rabinovitch A, Cornbleet PJ for the Hematology/Clinical Microscopy Resource Committee of the College of American Pathologists: Physician review of the peripheral blood smear—when and why. Lab Hematol. 2001;7:175–179.

Robert W. Novak, MD
Department of Pathology
Children’s Hospital
Medical Center of Akron (Ohio)
Chair, CAP Hematology/Clinical Microscopy Resource Committee