Early recognition and treatment of circulatory volume loss are essential in the clinical management of dengue viral infection. We hypothesized that a novel computational algorithm, originally developed for noninvasive monitoring of blood loss in combat casualties, could: (1) indicate the central volume status of children with dengue during the early stages of “shock”; and (2) track fluid resuscitation status.
Continuous noninvasive photoplethysmographic waveforms were collected over a 5-month period from three children of Thai ethnicity with clinical suspicion of dengue. Waveform data were processed by the algorithm to calculate each child’s Compensatory Reserve Index, where 1 represents supine normovolemia and 0 represents the circulatory volume at which hemodynamic decompensation occurs. Values between 1 and 0 indicate the proportion of reserve remaining before hemodynamic decompensation.
This case report describes a 7-year-old Thai boy, another 7-year-old Thai boy, and a 9-year-old Thai boy who exhibited signs and symptoms of dengue shock syndrome; all the children had secondary dengue virus infections, documented by serology and reverse transcriptase-polymerase chain reaction. The three boys experienced substantial plasma leakage demonstrated by pleural effusion index >25, ascites, and >20 % hemoconcentration. They received fluid administered intravenously; one received a blood transfusion. All three boys showed a significantly low initial Compensatory Reserve Index (≥0.20), indicating a clinical diagnosis of “near shock”. Following 5 days with fluid resuscitation treatment, their Compensatory Reserve Index increased towards “normovolemia” (that is, Compensatory Reserve Index >0.75).
The results from these cases demonstrate a new variation in the diagnostic capability to manage patients with dengue shock syndrome. The findings shed new light on a method that can avoid possible adverse effects of shock by noninvasive measurement of a patient’s compensatory reserve rather than standard vital signs or invasive diagnostic methods.
Nearly 400 million people are infected with dengue virus (DENV) every year, making dengue the most common mosquito-borne viral disease in humans . The clinical manifestations of DENV infection range from asymptomatic infection to a debilitating but self-limited illness termed dengue fever (DF), to more severe forms of the disease characterized by increased vascular permeability and plasma leakage, known as dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS) [2, 3]. Approximately 5 to 10 % of dengue patients experience DHF . The characteristic signs and symptoms of DHF may include abdominal pain, persistent vomiting, petechiae, epistaxis, gingival and gastrointestinal bleeding, microscopic hematuria, thrombocytopenia and, most importantly, hemoconcentration due to increased vascular permeability and third space fluid losses into the pleural and peritoneal cavities. Patients with DHF may progress to develop DSS, evidenced by signs of circulatory failure, including tachycardia, a narrow pulse pressure, hypotension and, in its late stages, lethargy and other mental status changes. Fatality rates among patients with DSS can be 10 % or higher. Fluid resuscitation is essential in DHF/DSS, but overly aggressive fluid administration can exacerbate third space fluid losses, which can lead to pulmonary edema and death when plasma reabsorption occurs. Fatality rates among patients with DSS are typically around 2.5 %, or approximately 22,000 deaths per year [4, 5]. There are no specific antiviral treatments for dengue.
Intravenous fluid therapy in DHF/DSS is guided by repeated physical examinations, frequent vital sign reassessment, serial hematocrits and imaging to assess for intravascular fluid losses and extravascular fluid accumulation (for example, pleural effusion, ascites). The latter strategies are invasive (blood draws) and include exposure to ionizing radiation (chest X-rays), both of which are undesirable in children. Although these strategies help to track volume status, compensatory mechanisms limit their sensitivity until patients are at or near the point of hemodynamic decompensation. Consequently, the capability to measure the integrated response of physiological mechanisms that reflect the sum total of compensation to a relative blood volume deficit could dramatically alter the monitoring of patients with dengue and multiple other conditions of hypovolemia. As such, the purpose of the case studies on patients with DHF reported in this paper was to determine if a novel computational algorithm [6–8], originally developed for detecting and monitoring blood loss in combat casualties, could noninvasively indicate the clinical status of children with dengue during the early stages of “shock”, and track their fluid resuscitation status.
The cases of a 7-year-old Thai boy, another 7-year-old Thai boy, and a 9-year-old Thai boy were selected to report in this paper. The children presented to the Queen Sirikit National Institute of Child Health (QSNICH), Bangkok, Thailand with clinical suspicion of dengue. None of the three boys met exclusion criteria that included any known chronic condition (for example, liver or renal disease, malignancy, or thalassemia). Informed written consent from a parent or guardian was obtained. This study was approved by the Institutional Review Boards of the QSNICH, Thai Ministry of Public Health, and US Army Surgeon General.
Our patients were classified and managed according to the World Health Organization (WHO) 1997 guidelines [2, 9, 10]. The intravenous fluid was initiated by the treating physician at QSNICH when one or more of the following were present: (1) signs suggestive of plasma leakage, that is, rising hematocrit ≥10 % concurrent with thrombocytopenia (platelet count ≤100,000/mm3) with poor oral intake; (2) signs of poor peripheral perfusion, including persistent tachycardia, delayed capillary refill (more than 2 seconds), or narrow pulse pressure (≤20 mmHg); and/or (3) need for blood or colloid solution transfusion determined by hematocrit and/or response to fluids administered intravenously.
The following clinical parameters were assessed daily during hospitalization: vital signs (oral temperature, respiratory rate, pulse, and blood pressure collected manually every 3 hours); physical examination findings (presence of ascites, liver enlargement, pleural effusion(s), and hemorrhagic manifestations); tourniquet test result; and complete blood count and serum albumin. Oral intake of fluid and intake of fluid administered intravenously and urinary output were recorded and totaled every 24 hours. A daily ultrasound examination was performed to assess for ascites, which was read as either present or absent. A chest X-ray was obtained on the day after defervescence (fever day +1) to assess for the presence of pleural effusion, measured as pleural effusion index . When a patient’s temperature was below 38 °C for two consecutive 6-hour periods, then hematocrit determination was conducted every 6 hours by finger stick until stable. In order to track the progression of illness around defervescence and the response to treatment, noninvasive blood pressure waveform data were collected . Patients were hospitalized until they were afebrile, had stable vital signs, and were able to tolerate oral feedings. Additional blood for diagnostic testing was collected on the day of enrollment and approximately 5 to 9 days after discharge.
Dengue cases were classified into DF or DHF grades I, II, III, or IV, based on the 1997 WHO case definition [2, 9, 10]. The final case classification of DSS (DHF >III) was retrospectively determined by a physician who was a dengue expert; the dengue expert reviewed the medical records but did not participate in inpatient care.
DENV in plasma collected at study entry was detected by a serotype-specific reverse transcriptase-polymerase chain reaction (RT-PCR) . Cases were classified as having primary or secondary DENV infection based on the ratio of dengue-specific immunoglobulin G and immunoglobulin M. and by haemagglutination inhibition assay on paired acute and convalescent samples as previously published . The percent hematocrit change was calculated as (highest hematocrit during hospitalization – hematocrit at convalescence)/(hematocrit at convalescence × 100).
Noninvasive blood pressure waveform data collection
Continuous noninvasive photoplethysmographic (PPG) waveforms were collected for approximately 15 minutes per day using a Nexfin blood pressure monitor (Edwards Lifesciences, Irvine, CA, USA). The sampling rate was 1000 Hz, exported at a rate of 200 Hz to a computer-based data acquisition software package (WinDAQ; Data Instruments, Akron, OH, USA).
The Compensatory Reserve Index (CRI) algorithm
The compensatory reserve represents a new paradigm for measuring the sum total of all compensatory mechanisms (for example, tachycardia, vasoconstriction, breathing) that together contribute to “protect” against inadequate tissue perfusion during blood loss and other low circulating blood volume states . The physiology is based on the premise that the sum total of all compensatory mechanisms involved in the control of cardiac output (ejected wave) and peripheral vascular resistance (reflected wave) are represented by the entirety of features of the total arterial waveform. Thus, alterations in specific waveform features represent changes in the reserve capacity to compensate for reduced circulating blood volume. This concept is illustrated by the changes in the arterial waveform between “normovolemia” and central “hypovolemia” presented in Fig. 1. Feature extraction and state-of-the-art machine-learning techniques were used to develop a novel computational algorithm that is capable of identifying and monitoring patients during the compensatory phase of central blood volume loss [6–8]. The algorithm was developed at the University of Colorado using data obtained from experiments conducted at the US Army Institute of Surgical Research . In these experiments, lower body negative pressure (LBNP) was used to induce carefully controlled reductions in central blood volume in 100 human participants and validated on 101 different participants . The way the Compensatory Reserve Index (CRI) algorithm for measuring compensatory reserve works is by processing 100 million data points per second in order to evaluate subtle changes in hundreds of specific features of the arterial waveform (outlined in red in Fig. 1). The algorithm used for the current study analyzes how a select group of noninvasive blood pressure waveform features change over time, from normovolemia to decompensation. By monitoring multiple waveform features and knowing how these features change with central blood volume loss, the algorithm determines how near or far an individual may be from the point of decompensation. The algorithm outputs a single value in a beat-to-beat fashion, termed the CRI . Values between 1 and 0 indicate the compensatory reserve of the patient or the proportion of reserve capacity that remains to compensate for central blood volume loss before the onset of decompensation at a CRI of zero.