This Article Abstract Full Text (PDF) All Versions of this Article: 90/8/4555    most recent Author Manuscript (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Service, F. J. Articles by O’Brien, P. C. Search for Related Content PubMed PubMed Citation Articles by Service, F. J. Articles by O’Brien, P. C. Related Collections Endocrine Oncology Metabolism

Increasing Serum Betahydroxybutyrate Concentrations during the 72-Hour Fast: Evidence against Hyperinsulinemic Hypoglycemia

Division of Endocrinology, Metabolism and Nutrition (F.J.S.) and Department of Biostatistics (P.C.O.), Mayo Clinic, Rochester, Minnesota 55905

Address all correspondence and requests for reprints to: F. John Service, M.D., Ph.D., Mayo Clinic, Division of Endocrinology, Metabolism and Nutrition, 200 First Street SW, Rochester, Minnesota 55905. E-mail: service.john{at}mayo.edu.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Objective: We have determined whether the behavior of betahydroxybutyrate (BOHB) during the 72-h fast of persons without evidence of hyperinsulinemic or any form of organic hypoglycemia might provide indicators of a negative fast.

Design: Twenty-one patients with surgically confirmed insulinoma and 34 patients with negative 72-h fasts had BOHB measured every 6 h until Whipple’s triad in the former and until 72-h in the latter.

Results: Quadratic regression curves of BOHB from the negative fasts showed the typical curve to be flat initially, then increase in a manner that was roughly linear. Using time-specific medians, the changes were: 12–18 h, 0% increase; 18–36 h, 333% increase; 36–54 h, 210% increase, and 54–72 h, 167% increase. In contrast, patients with insulinoma had suppressed BOHB concentrations. Two successive BOHB values in excess of the 18-h level seemed to portend a negative fast. By using the previously published criterion of BOHB more than 2.7 mmol/liter (a surrogate for hypoinsulinemia and thereby an indicator of a negative fast), 74% of persons with a negative fast reached this level before the 72-h point.

Conclusions: Serial measurements of BOHB during the 72-h fast have the potential to provide not only clues during the fast that it will ultimately be negative, but also the opportunity to truncate the fast if the endpoint BOHB criterion for a negative fast is met before 72 h.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
HYPOGLYCEMIC DISORDERS ARE usually manifested by episodic rather than sustained hypoglycemia. Documentation of hypoglycemia during a spontaneous episode may be difficult because of the unpredictable occurrence of hypoglycemic events; in such a situation, provocative testing is needed. The 72-h fast is the established provocative test for hypoglycemia occurring in the food-deprived state (1). Based on the widely accepted experience from patients with insulinoma who, for the most part, show positive results within a short period of fasting, a recommendation has been proffered to truncate the 72-h fast at 48 h (2). Our experience of a 93% positivity rate within 48 h of fasting and a 99% positivity rate by 72 h of fasting of patients with insulinoma has convinced us not to abandon the 72-h fast in favor of the 48-h fast (1). An additional factor in the conduct of the 72-h fast overlooked by the proponents of the 48-h fast is the need to determine whether a 72-h fast is indeed negative. The lack of a positive result by 48 h cannot be deemed a declaration of a negative fast. Furthermore, a negative 72-h fast is not the default interpretation of a nondiagnostic fast, but indeed is a definite diagnostic entity (3, 4).

From time to time, over more than three decades of conducting the 72-h fast in scores of persons suspected to have a hypoglycemic disorder, we have encountered some who developed symptomatic ketosis as a result of accelerated ketogenesis secondary to hypoinsulinemia well before the 72 h had elapsed. Whereas, with encouragement, most of these individuals completed the fast, some declined. Among the latter, the concentrations of betahydroxybutyrate (BOHB), an insulin surrogate, exceeded the diagnostic criterion (>2.7 mmol/liter) for a negative 72-h fast, despite having had an abbreviated fast (3). Therefore, we reasoned that changes in BOHB in the 72-h fast might presage a negative fast.

In 2002, we modified our standard 72-h fast protocol (3) by adding serial measurements of BOHB primarily for the safety reason of identifying symptomatic ketosis. During the subsequent period, 34 patients had negative 72-h fast and 21 patients with surgically confirmed insulinoma had positive 72-h fasts. A negative 72-h fast was determined by absence of hypoglycemic symptoms and a terminal serum glucose more than or equal to 60 mg/dl or a terminal serum glucose less than 60 mg/dl with suppressed concentrations of insulin (<3 µU/ml), C-peptide (<200 pmol/liter) and proinsulin (<5 pmol/liter) (3, 4). A retrospective analysis with waived consent on the basis of minimal risk was conducted with Institutional Review Board (IRB 1083-04) approval.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Plasma insulin (5), C-peptide (6), and proinsulin (7) were measured by immunochemiluminometric assays with lower limits of detection of 0.1 µU/ml and 33 pmol/liter and 0.6 pmol/liter, respectively. Plasma sulfonylureas were measured by liquid chromatographic tandem mass spectroscopy (8, 9). Serum glucose was measured by standard hexokinase method on a Hitachi 747-200 Chemistry Analyzer (Roche Diagnostics, Indianapolis, IN). BOHB was measured by an automated kinetic method (10).

Statistical analysis

Differences between groups: positive vs. negative 72-h fast and more than or equal to 60 vs. less than 60 mg/dl glucose in negative 72-h fasts was performed using Wilcoxon rank sum test. Inspection of the data revealed progressively increasing levels of BOHB over time among the patients with negative 72-h fasts, whereas BOHB levels remained flat in those patients with positive 72-h fasts. Specifically, the occurrence of two consecutive values of BOHB which exceeded the 18-h value appeared to characterize the curves observed in the patients with negative 72-h fasts (Fig. 1). In an effort to further characterize the profiles observed among those with negative 72-h fasts additional regression analyses were conducted. Regression equations were obtained for each subject, then a mean curve was obtained by computing the means of each of the regression coefficients. Because the data were highly skewed, a log transformation was used.

View larger version (11K): [in this window] [in a new window]   FIG. 1. Median concentrations of serum BOHB during 72-h fasts are shown for patients with insulinoma, all of whom had positive fasts, and for patients who had negative fasts.

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

Among the 34 patients with negative 72-h fasts and the 21 patients with positive 72-h fasts, there were no differences in age or gender, but body mass index (BMI) was greater in the latter (Table 1). Those with positive fasts had a shorter duration of fast and lower serum glucose and BOHB concentrations (Table 1). Among the patients with negative 72-h fasts, there were no differences in age, BMI, nor duration of fast for those with terminal serum glucose more than or equal to 60 and less than 60 mg/dl. There was a significant gender difference between the two groups with female predominance and higher BOHB concentrations at the termination of the fast in those with terminal glucose less than 60 mg/dl (Table 2). This group had suppressed concentrations of plasma insulin below our established criterion for hyperinsulinemia in all patients (1.0, 0.5–2.0 µU/ml) and of C-peptide (132, 33–364 pmol/liter) in 17 of 20 patients and of proinsulin (4.3, 2.2–8.3 pmol/liter) in 15 of 20 patients. Those without completely suppressed C-peptide and proinsulin concentrations had higher terminal plasma glucose concentrations consonant with our previous observations (4). All patients had negative serum sulfonylurea screen and no detectable insulin antibodies.

The quadratic regression curve for the negative fasts, log BOHB = –2.42 + 0.0874(time) + 0.0005(time-squared), demonstrated that a typical curve was initially flat with time, then increased in a manner which was roughly linear.

This phenomenon is seen in the time-specific medians: 12–18 h, 0% increase; 18–36 h, 333% increase; 36–54 h, 210% increase, and 54–72 h, 167% increases (Fig. 1).

For the 21 patients who had negative 72-h fasts and complete data sets permitting assessment of the occurrence of two successive increased BOHB levels above the 18-h value, concentration of BOHB at the 18-h point and at the time of the first and second successive increase in BOHB and their timing are shown in Table 3. Once two consecutive values of BOHB above the 18-h value were observed, no subsequent value was found below the 18-h level. Among the 21 patients—14 in the less than 60 mg/dl group and seven in the greater than or equal to 60 mg/dl group—14 demonstrated the second consecutive increased BOHB at the earliest possible time point, 30 h; five at 36 h, and the other two at 42 and 48 h. There was no difference in time to attainment of the second consecutive increase in BOHB between the less than 60 and more than or equal to 60 mg/dl glucose groups (Table 3). The attainment of two successive BOHB concentrations more than 2.7 mmol/liter before the 72-h point occurred in 19 of 20 patients in the less than 60 mg/dl group at 54, 36–66 h and in 6 of 14 patients in the more than or equal to 60 mg/dl group at 60, 12–66 h, respectively. The attainment of two successive values of BOHB of 2.0, 2.5, and 3.0 mmol/liter occurred in the less than 60 mg/dl group in 19 of 20 patients at 48, 30–66 h; 54, 36–66; and 63, 36–72 h. Five patients in the more than or equal to 60 mg/dl group did not reach the minimum level of 2.0 mmol/liter. Among the others in this group, the time to reach each of these levels of BOHB did not differ from the less than 60 mg/dl group In none of the insulinoma patients in this study did the BOHB reach 1.0 mmol/liter.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

The strength of our observation of an early and sustained increase in BOHB as a possible indicator of a negative 72-h fast would be reinforced if all of the patients with a positive fast could have sustained the fasting period to 72 h. In actuality, it is uncommon for 72-h fasts in patients with fasting state hypoglycemia such as insulinoma to proceed to 72 h. In a previously reported series of 72-h fasts in 170 patients with insulinoma, 65% were positive by 24 h (1). In the insulinoma patients in this study, 85% were positive by 24 h. Hence, any patient still fasting beyond 24 h has a progressively higher likelihood of having a negative fast. Accumulation of experience with insulinoma patients fasted into the third 24-h period of the 72-h fast for comparison with negative fasts would likely take many years to accrue even in a tertiary center with considerable experience with this disorder. In lieu of these data, it appears entirely reasonable that once the BOHB has reached a sustained level of more than 2.7 mmol/liter the fast could be stopped and declared negative. Of course, in any patient where the suspicion of a positive fast is high, the fast should be continued to 72 h or until there is diagnostic evidence for a positive fast.

	Footnotes

 
First Published Online May 10, 2005

Abbreviations: BMI, body mass index; BOHB, betahydroxybutyrate.

Received January 7, 2005.

Accepted May 3, 2005.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Service FJ, Natt N 2000 Clinical perspective: the prolonged fast. J Clin Endocrinol Metab 85:3973–3974[Free Full Text]
2. Hirshberg B, Livi A, Bartlett DL, Libutti SK, Alexander HR, Doppman JL, Skarulis MC, Gorden P 2000 Forty-eight hour fast: the diagnostic test for insulinoma. J Clin Endocrinol Metab 85:3222–3226[Abstract/Free Full Text]
3. Service FJ 1995 Hypoglycemic disorders. N Engl J Med 332:1144–1152[Free Full Text]
4. Service FJ 1999 Diagnostic approach to adults with hypoglycemic disorders. Endocrinol Metab Clin North Am 28:519–532[CrossRef][Medline]
5. Threatte GA, Henry JB 1996 Carbohydrates. In: Clinical diagnosis and management of laboratory methods. Henry JB, ed. 19th ed. Philadelphia: WB Saunders
6. Kao PC, Taylor RI, Heser DW 1992 C-peptide immunochemilunometric assay developed from two seemingly identical polyclonal sera. Ann Clin Lab Sci 22:307–316[Abstract]
7. Kao PC, Taylor RL, Service FJ 1994 Proinsulin by immunochemiluminometric assay for the diagnosis of insulinoma. J Clin Endocrinol Metab 78:1048–1051[Abstract]
8. Magni F, Marazzini L, Pereira S, Monti L, Galli KM 2000 Identification of sulfonylureas in serum by electrospray mass spectrometry. Anal Biochem 282:136–141[CrossRef][Medline]
9. Sacks DB 1999 Carbohydrates. In: Burtis CA, Ashwood ER, eds. Tietz textbook of clinical chemistry. Philadelphia: WB Saunders
10. McMurray CH, Blanchflower JW, Rice DA 1984 Automated kinetic method for D-3-hydroxybutyrate in plasma or serum. Clin Chem 30:421–425[Abstract]
11. Merimee TJ, Fineberg SE 1973 Homeostasis during fasting II. Hormone substrate differences between men and women. J Clin Endocrinol Metab 37:698–702[Medline]
12. O’Brien T, O’Brien PC, Service FJ 1993 Insulin surrogates in insulinoma. J Clin Endocrinol Metab 77:448–451[Abstract]

This article has been cited by other articles:

				D Vezzosi, A Bennet, J Fauvel, and P Caron Insulin, C-peptide and proinsulin for the biochemical diagnosis of hypoglycaemia related to endogenous hyperinsulinism Eur. J. Endocrinol., July 1, 2007; 157(1): 75 - 83. [Abstract] [Full Text] [PDF]

				P. Kar, P. Price, S. Sawers, S. Bhattacharya, R. H. Reznek, and A. B. Grossman Insulinomas May Present with Normoglycemia after Prolonged Fasting but Glucose-Stimulated Hypoglycemia J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 4733 - 4736. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 90/8/4555    most recent Author Manuscript (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Service, F. J. Articles by O’Brien, P. C. Search for Related Content PubMed PubMed Citation Articles by Service, F. J. Articles by O’Brien, P. C. Related Collections Endocrine Oncology Metabolism