Although, according to the most recent World Malaria report, there were still an estimated 655,000 deaths from the disease in 2010, the majority occurring in African children, major investments in malaria control by the international community in the past decade have yielded excellent returns, with mortality rates falling by more than 25% globally. However as we approach this year’s World Malaria Day (25th April), it may be prudent to consider widening the focus of these international control efforts.
Of the five distinct species of Plasmodium causing human malaria, two species, namely P. malariae and P. ovale, have low prevalence and normally only cause mild disease. The thrust of global control efforts has been directed at P. falciparum, one of the two highly prevalent species, because it causes the highest mortality. However the former name of ‘benign tertian malaria’ given to P. vivax, the other highly prevalent species that is also the most widely distributed, is certainly a misnomer; infection with P. vivax is anything but mild. It has been estimated that this species, endemic in South and Central America, the Middle East, Africa and Asia, and found in temperate as well as tropical areas, causes over two hundred million cases of malaria per year. While fatal infections, usually resulting from a ruptured spleen, are infrequent, P. vivax is harder to diagnose than P. falciparum as it infects immature red blood cells and parasitaemias are thus lower. It is also harder to treat because the life cycle includes dormant liver stages (hypnozoites) that cause periodic relapse infections, accompanied by severe anaemia, respiratory distress and poor obstetric outcomes. And it affects all age groups rather than predominantly children, so as well as the human suffering endured, the economic impact is huge.
The fifth species of Plasmodium that can cause human malaria, P. knowlesi, was previously only thought to infect certain species of macaque monkeys, but has now been recognised as a clinically significant zoonosis. It has been reported from several South East Asian countries, including Thailand, Malaysia, Vietnam, Myanmar, Singapore, Indonesia and the Philippines, and causes up to 70% of the malaria cases in some of these areas. As with P. falciparum, infection with P. knowlesi is potentially fatal if it is not diagnosed and treated promptly; unfortunately microscopically it is very similar to the much less serious P. malariae and is frequently misdiagnosed. And a major concern is that deforestation and increasing human settlement in P. knowlesi endemic areas may result in humans, rather than macaques, becoming the preferred host, and thus the dissemination of P. knowlesi to neighbouring countries where there are no suitable simian hosts, but where the vector mosquitoes (predominantly Anopheles leucophyrus group) breed.
International investment and efforts to control malaria in the last decade have been truly laudable, but it is now time to look outside the P. falciparum box.

ß-Hydroxybutyrate (BHB) is the main ketone body produced during ketosis and diabetic ketosis. This article demonstrates how quantitatively measuring plasma/serum BHB levels can give a direct indication of blood ketone levels and provide a more accurate method of diagnosing and managing ketosis than via traditional nitroprusside-based urine dipstick testing. Rapid identification of ketosis through BHB testing can improve clinical management and patient care in diabetes.

by Dr Cormac Kilty and Al Blanco

What is Ketosis?
Ketosis occurs when the body begins to break down its stored fats in response to a low supply of energy (glucose) to produce ketone bodies. These water soluble by-products of fatty acid metabolism are then used by the body as alternative energy sources to reduce tissue demand for glucose.

Ketone bodies are always present in the blood and are normally broken down into carbon dioxide and water. However, ketone build-up in the blood (ketonemia) can result from both physiological and pathological causes. Physiological ketosis, leading to a mild to moderate build-up, can result from prolonged exercise, fasting or a high-fat diet. If the cause is pathological, then ultimately the excessive build-up of ketones causes an acid/base imbalance known as ketoacidosis.  Pathological causes of ketosis include: diabetes mellitus, alcoholism, glycogen storage disease, alkalosis, ingestion of isopropyl alcohol, and salicylate poisoning. If not diagnosed and treated, ketoacidosis is potentially fatal.

The ketone bodies produced during ketosis within the liver are ß-Hydroxybutyrate (BHB), acetoacetate (AcAc) and acetone, where BHB is the predominant ketone body (78%) which is metabolized from AcAc [Figure 1]. The ketone body ratio, which is the ratio of BHB to AcAc, is approximately 1:1 in healthy people, but this can rise to nearly 6:1 after prolonged fasting and even 10:1 in cases of acute pathological ketosis [1].

Diabetic Ketoacidosis
Pathological ketosis most commonly arises due to diabetes mellitus (DM), a metabolic disease resulting in chronically high blood sugar. This occurs due to glucose under-utilisation and over-production in response to either: 1) an inability to produce and secrete insulin (Type 1 diabetes), or 2) insulin resistance (Type 2 diabetes) [2].

Diabetic ketoacidosis (DKA) is a life threatening complication of untreated or poorly managed diabetes which is most typically seen in the setting of Type 1 diabetes; in these cases the lack of insulin prevents the body from utilizing glucose for energy. This is because insulin acts on cell receptors to assist with glucose absorption, so in its absence cells are unable to take in, and subsequently metabolize, glucose. When the body senses glucose is not readily available, DKA occurs as fat is broken down instead. Furthermore, blood glucose levels rise (usually higher than 300 mg/dL) due to the over production of glucose by the liver to try to compensate for the problem. However, this additional glucose also cannot be metabolized without insulin [Figure 2] resulting in hyperglycemia. Although more common in patients with Type 1 diabetes, patients with Type 2 diabetes are also at risk of developing DKA during catabolic stress in the setting of trauma, surgery or infection [5]. There is also a subset of patients with Type 2 diabetes who are prone to ketosis.  They present with transient and severe beta cell dysfunction and the clinical course is variable. 

Rapid diagnosis of DKA is essential because a delay in starting insulin treatment is associated with an increase in morbidity and mortality [4]. Before insulin treatment was available, DKA was once the leading cause of death among Type 1 diabetics. Even now there is still a high mortality rate of 5 to 10% in developed countries, and it is the leading cause of death in pediatrics and young adults [5].

Testing for DKA
Like glucose, ketones can be tested or monitored in either urine or blood. Historically, DKA has been identified using a colorimetric semi-quantitative method for detection of ketones in urine. Nitroprusside turns purple in the presence of acetoacetate (Figure 3).  Although it is simple and rapid the dipstick nitroprusside test has several limitations, primarily that it only measures acetoacetate. False positives can result from interference by drugs such as L-Dopa, Captopril and other ACE inhibitors. False negatives can also occur because nitroprusside does not test for the presence of BHB which is the predominant ketone in DKA (>0.27 mmol/L is abnormal). Consequently, tests that only recognise the presence of AcAc will underestimate the total ketone body concentration (6). Furthermore, monitoring AcAc levels by using nitroprusside testing during DKA treatment can be misleading because a patient in DKA converts BHB to AcAc and acetone with insulin treatment. Therefore, nitroprusside tests will have a stronger reaction than prior to treatment, even though ketoacidosis is actually improving. This is because the fall in AcAc lags behind the improvement in ketoacidosis. By monitoring BHB levels instead, clinicians are able to assess the patient’s direct response to DKA treatment and ascertain immediately when ketoacidosis is resolved.

There are several different methods of testing for BHB in blood, plasma or serum, these include gas chromatography and capillary electrophoresis.  Such methodologies are specific, but they are more complex procedures that are not amenable to all hospital laboratory or clinic testing. In addition, the turnaround time can be longer than the one to two minutes of the nitroprusside method.

Rapid β-Hydroxybutyrate measurement
An enzymatic assay is also available for direct quantitative measurement of BHB in blood. This is rapid, has minimal cross reactivity with interfering substances and can be performed on both automated laboratory instrumentation (for plasma/serum), or using whole blood samples on devices at the point of care.  An example of this assay is presented by the β-Hydroxybutyrate LiquiColor® Reagent System (Stanbio, Boerne, TX, USA). Figure 4 details the enzymatic reaction which gives a purple colour proportional to the concentration of BHB; where normal levels are 0 – 0.3 mM/L, ketosis is >0.3 mM/L and possible ketoacidosis is >5 mM/L.

Recent prospective studies have shown that blood BHB enzymatic testing has a far superior specificity in comparison to the nitroprusside urine test [7].  One such study prospectively screened for DKA in emergency department (ED) patients who had a blood sugar of >250 mg/dL, regardless of the reason for the ED visit. Both a urine dipstick and a point of care capillary BHB test were performed, with both tests displaying an acceptable sensitivity of at least 98%. However, the BHB was markedly more specific at 78.6%, in comparison to the urine dipstick (35.1% specificity) [7].  The American Diabetes Association discourages the use of urine nitroprusside testing and instead recommends quantitative serum BHB testing for diagnosing and monitoring ketoacidosis [8]. Furthermore, the Association recommends that blood ketone determinations that rely on the nitroprusside reaction should only be used as an adjunct to diagnose DKA and should not be used to monitor DKA treatment due to the lag in decrease in AcAc after resolution of ketoacidosis. In contrast, specific measurement of BHB in blood can be used for diagnosis and monitoring of DKA [9, 10].

Clinical advantages of BHB
As BHB testing is rapid and more specific than urine nitroprusside testing for ketones; it can be used to identify ketosis in multiple settings. BHB in serum and plasma can be used to clinically diagnose and monitor the disease status and severity of diabetes mellitus, alcoholism, as well as starvation-induced ketosis. It may also have potential application for diagnosing and monitoring glycogen storage disease, high fat/low carbohydrate diets, ingestion of isopropyl alcohol and salicylate poisoning. BHB testing is rapid and more specific than urine nitroprusside testing for ketones since it tests for the main ketone produced during ketosis (78%).

During ketosis, BHB levels increase more than the levels of acetone and acetoacetate, clearly indicating the patient’s trend in metabolic status. Consequently, quantitative, objective BHB results provide a better tool for determining and monitoring ketosis than qualitative nitroprusside testing that detects only 22% of ketones present during ketosis.

BHB testing gives the earliest detection of clinically significant ketosis, enabling clinicians to diagnose DKA with confidence based on quantifiable results. Rapid identification of ketosis through BHB testing can improve clinical management and patient care [4]. As such, early detection could enable shorter triage times and faster treatment of patients which could in turn lead to improved clinical outcomes and Emergency Department efficiency, and decreased turnaround times [11]. Furthermore, unnecessary patient admissions could also be avoided through faster and more accurate patient assessments for ketosis and ketoacidosis, particularly within the Emergency Department, which may also result in ever important cost savings.

Acknowledgement
The authors thank Dr. James H. Nichols, Ph.D., DABCC, FACB, Professor of Pathology, Tufts University School of Medicine and Medical Director for Clinical Chemistry at Baystate Health in Springfield, MA. This manuscript is based on a presentation given by Dr. Nichols at the July, 2012 American Association for Clinical Chemistry meeting in Los Angeles.

References
1. Laffel L. Diabetes/Metabolism Research and Reviews 1999; 15:412-426.
2. Shoback, edited by David G. Gardner, Dolores (2011). Greenspan’s basic & clinical endocrinology (9th ed. ed.). New York: McGraw-Hill Medical. pp. Chapter 17
3. Kitabchi AE, et al. Diabetes Care 2009; 32(7):1335-1343.
4. Singh RK, et al. Diabet Med 1997;14:482-486.
5. Felner E, et al. Pediatrics 2001;108:735-740.
6. Sacks DB, et al.  Diabetes Care 2004:34:e61-e99.
7. Arora S, et al. Diabetes Care. 2011; 34(4):852-4.
8. American Diabetes Association. Diabetes Care 2010; 33 (Suppl 1); S62-69.
9. Sacks DB, et al.  Diabetes Care 2011; 34:1419-1423
10. Savage MW, et al. Diabetic Medicine 2011; 28(5):508-515.
11. Foreback C, Former Director of Clinical Chemistry, Henry Ford Hospital, Detroit, MI, White Paper, Clinical effectiveness of Beta-Hydroxybutryate assays in a clinical decision unit, (1998).

The authors
Dr Cormac Kilty
EKF Diagnostics Holdings plc, UK
Tel. +44 (0)2920 710 570
E-mail: cormackilty@ekfdiagnostics.com

Al Blanco
Stanbio Laboratory (An EKF Diagnostics company), USA
Tel. +1 (0)830 249 0772