Ask The Experts

Diagnosing Diabetes Subtypes: Copeptin or Water Deprivation Test?

Blood testing for copeptin may offer an easier diagnostic method than the customary water-deprivation in assessing for diabetes insipidus.

With Gary Robertson, MD, and Mirjam Christ-Crain, MD

An alternative to the indirect water deprivation test—the current reference standard—may be more accurate for diagnosing the rare disorder of diabetes insipidus, 1 according to findings published in the New England Journal of Medicine.

Mirjam Christ-Crain, MD, of University Hospital Basel in Switzerland and her team performed both the copeptin test and the water deprivation test on 144 patients to evaluate the accuracy of these diagnostic approaches.1

All patients had confirmed hypotonia polyuria and were receiving care across 11 medical centers. The primary outcome of this study was to compare the overall diagnostic accuracy of each test compared with the final reference diagnosis.1 The participants were given both water- and hypertonic saline infusion tests. In the latter, the plasma copeptin was measured when the plasma sodium level had increased to at least 150 mmol per liter after infusion of hypertonic saline.
Testing 24-hour urine followed by blood test is best method for diagnosing diabetes insipidus subtype.

"The most important finding is that a new test, the hypertonic saline infusion copeptin measurement, has a much higher diagnostic accuracy than the classical water deprivation test in the differential diagnosis of polyuria polydipsia syndrome, and is also well received by the patients, even better than the water deprivation test," Dr. Christ-Crain told EndocrineWeb, suggesting that the copeptin approach become the new ''gold standard" for diagnosing the cause of hypotonic polyuria. However, not everyone agreed.

The water deprivation test induced one case of desmopressin-induced hyponatremia requiring hospitalization.

Assessing Diagnostic Approach in Patients with Hypotonic Polyuria

A final diagnosis was determined after evaluating medical history, test results from both approaches, and treatment response with the copeptin levels masked.1 The researchers reported that 82 patients (57%) received a diagnosis of primary polydipsia, 59 patients (41%) were confirmed as having central diabetes insipidus and three patients were diagnosed with nephrogenic diabetes insipidus.

Among the 141 patients in the final analysis:1

  • Water-deprivation test determined the correct diagnosis in 108 (accuracy, 76.6%, 95% CI, 68.9-83.2)
  • Hypertonic saline infusion test (with a copeptin cutoff of > 4.9 mmol per liter) determined the correct diagnosis in 136 (accuracy 96.5%, 95% CI, 92.1-98.6, P > 0.001.)
  • Water test correctly distinguished primary polydipsia from partial central diabetes insipidus in 77 of 105 (73.3%; 95% CI, 63.9-81.2).
  • Hypertonic saline infusion distinguished between the two in 99 of 104 (95.2%, 95% CI, 89.4-98.1, adjusted P < 0.001).

While Dr. Christ-Crain advocated in favor of the new test, she cited one caveat:  "[The copeptin] test may lead to a clear osmotic stimulation with sodium levels of 150 mmol/L. This change may induce adverse effects in some patients (ie, thirst, headache, nausea), although these symptoms were in general rather mild, and there was no serious adverse effect," she said.

However, in stimulating copeptin, these saline infusions could, for instance, induce congestive heart failure in high-risk patients, requiring constant surveillance during the procedure, she said. Sodium control has to be performed after the test is done with sodium lowering achieved with a glucose infusion and/or by having the patient drink of water.

Expert Overview of Diagnostic Testing for Diabetes Insipidus

"Contrary to the opinion expressed by the authors of this paper,1 I believe the most important message for physicians is to be highly skeptical of plasma copeptin assays to differentiate partial central (pituitary) diabetes insipidus (DI) from primary polydipsia,” said Gary Robertson, MD, professor emeritus of medicine at the Feinberg School of Medicine at Northwestern University, in Chicago, who reviewed the study for EndocrineWeb.

“Eventually, it may prove useful but when employed in the manner described by Crist-Crain, it may be much less reliable for differentiating between the three major types of diabetes insipidus than other existing methods such as the assay for basal plasma arginine vasopressin (AVP) with brain MRI when indicate,” Dr. Robertson said.

On the other hand, ''in agreement with previous reports of more than 40 years,2-4 [these authors] found that the traditional indirect water deprivation test was very reliable for differentiating severe pituitary (central) diabetes insipidus from other forms of this rare form of diabetes,” he said.

“Also, they reported that the traditional indirect fluid deprivation test was not as reliable for differentiating between partial pituitary diabetes insipidus—in which the deficiency of vasopressin-AVP—is less severe, and primary polydipsia, a disorder that mimics diabetes insipidus but arises from an excessive water intake.”

“The difficulty in diagnosing the cause of polyuria arises in their claim that this distinction can be made more accurately or easily by measuring plasma copeptin (a peptide larger than AVP and synthesized in the posterior pituitary with it) before and after hypertonic saline infusion than by other methods, such as measuring plasma AVP before and after 3% saline infusion or simply measuring basal plasma AVP and, if it is low, performing a brain MRI to determine the if the posterior pituitary bright spot is present or  absent,'' said Dr. Robertson.

Critical Shortcomings in the Protocol of this Study

Dr. Robertson took issue with several other points in the study, including the lab criteria used to make a diagnosis of diabetes insipidus; the urine osmolarity values in many participants would not meet the accepted definition of primary polydipsia nor of partial central (pituitary) diabetes insipidus.3

"Another major shortcoming was that the levels of plasma sodium achieved during 3% saline infusion appeared markedly different in the two groups. The percent in which plasma sodium exceeded 155 mmol/L was much higher in the subjects labeled primary polydipsia (19%) than those with a diagnosis of partial central (pituitary) DI (4%),” Dr. Robertson said. “That difference alone could easily produce differences in plasma copeptin and would be of little or no diagnostic value. To avoid errors of this kind, the levels of copeptin should be expressed relative to plasma sodium (or osmolarity) similar to the analysis of AVP.”

"The most significant defect in this study design is the lack of objective criteria to determine the accuracy of the diagnosis made by the assay of copeptin (the so-called “gold standard”), he told EndocrineWeb.

“Furthermore, the accuracy of the diagnosis was evaluated by comparing it to the opinion of outside 'experts' and clinical information other than the results of the copeptin assay. The authors do not provide any information about the response to desmopressin therapy, which is the best criterion and the primary reason for differentiating partial central (pituitary) DI from primary polydipsia,” he said.

“If prescribed in the correct amounts,5 desmopressin completely corrects the polyuria and polydipsia without producing hyponatremia (water intoxication) in patients with partial central (pituitary) DI but, in primary polydipsia, the same doses of desmopressin abolish the polyuria but not the polydipsia and, as a consequence, invariably result in symptomatic water intoxication,” said Dr. Robinson.

The bottom line, according to Dr. Robertson, is for clinicians ''to be wary of a diagnosis made using the copeptin method described.”

A More Reliable Method for Diagnosing the Subtype of Diabetes Insipidus

“The easiest and most reliable method for evaluating a patient with suspected diabetes insipidus is the following:2,3

Collect a 24-hour urine


If the volume is abnormally high (greater than 40 or 50 mL/kg body weight)
and osmolarity is low (less than 300 mOsm/L)


Measure basal plasma AVP or copeptin


If this measure is low or undetectable, the Diabetes Inspidius is likely not nephrogenic


Order MRI of the brain without contrast 
in order to determine the presence or absence of the normal pituitary.
A bright spot on T-1 weighted images will determine if the patient has:

·       Primary polydipsia—bright spot present

·       Central (pituitary) diabetes insipidus—bright spot small or absent3

The copeptin test, Dr. Robertson said, ''may be a reliable tool for differentiating nephrogenic diabetes insipidus from the other types of DI. Although the information on the reliability of this method is inadequate for full confidence, it may be easier to obtain and with proper improvements in the methods of interpretation may eventually prove to be a good alternative.''

Dr. Christ-Crain reports receiving personal fees from Thermofisher Scientific; Dr. Robertson has no financial disclosures relevant assessing this study.  

Continue Reading:
ADA Prioritizes Active Patient Engagement to Improve Diabetes Care
close X