Dengue in the elderly: a review

ABSTRACT

Introduction: Changing dengue epidemiological trends have resulted in a shift in the disease burden to the adult population. Older adults suffer from poorer outcomes as compared to their younger counterparts, making clinical management of this sub-population particularly challenging.

Areas covered: We present a review of the current literature on the changing epidemiology of dengue in the elderly, the atypical features of the clinical disease in this population with emphasis on severe disease presentations and challenges in the current management strategies.

Expert commentary: Dengue in the elderly is an increasingly important yet greatly understudied area. There is an urgent need to refine the current diagnostic criteria to improve diagnosis, classification of disease severity and identify individuals in this population who are likely to progress to severe disease. Management strategies in this population would have to be adjusted to account for the increased number of co-morbidities. The role of the available dengue vaccines in this group is uncertain, and more studies into their safety and efficacy are critically needed.

KEYWORDS:

• Dengue
• elderly
• clinical presentation
• diagnosis
• management

1. Introduction

Dengue is an important arthropod-borne viral infection which poses a global public health problem. An estimated 3.9 billion people in 128 countries are at risk of infection [1].and dengue virus infections account for nearly 500,000 hospitalizations annually [2]. The incidence of dengue has risen dramatically over the years, with the number of reported cases from member states of the World Health Organization (WHO) in South-East Asia, the Americas and the Western Pacific increasing from 2.2 million in 2010 to 3.2 million in 2015 [2].

Dengue is transmitted predominantly by the domestic Aedes aegypti mosquito, and to a lesser extent, the peridomestic Aedes albopictus mosquito [3]. Belonging to the Flaviviridae family, there are 4 distinct but closely related serotypes of dengue virus (DEN-1 to DEN-4). Infection with any one of the four dengue serotypes can result in a wide range of clinical manifestations, from a mild undifferentiated febrile illness, to classical dengue fever (DF), and to life-threatening severe dengue.

Classical DF is a clinical syndrome characterized by an abrupt onset of fever, which may be accompanied by chills, asthenia, headache, retro-orbital pain, backache, myalgia, arthralgia, anorexia, nausea and vomiting, diarrhea and a generalized maculopapular rash. Progressive leukopenia and thrombocytopenia are common during this febrile period [4]. Petechiae, gum bleeding, and epistaxis may also occur. Symptoms start 4–10 days after the bite from an infected mosquito and usually last for 2–7 days. Initial dengue virus infections are typically mild or subclinical, although cases of severe primary dengue have been reported [5]. Recovery from the infection provides lifelong immunity to that particular dengue virus serotype. However, cross-immunity to the other serotypes after recovery is only partial and temporary. Subsequent infections by other serotypes occur not infrequently, and is associated with an increased risk of developing severe dengue due to the phenomenon of antibody-dependent enhancement by non-neutralizing antibodies [6].

Severe dengue is a potentially fatal form that is characterized by plasma leakage leading to fluid accumulation, respiratory distress, severe bleeding, or organ impairment. Warning signs occur 3–7 days after the first symptoms in conjunction with the abatement of the febrile phase and manifest as severe abdominal pain, persistent vomiting, clinical fluid accumulation, bleeding from mucosal surfaces, lethargy, restlessness, liver enlargement or laboratory findings of a rising hematocrit concurrent with a rapid decrease in platelet counts [7]. Rapid deterioration, and even death, may ensue in the next 24–48 h if appropriate clinical intervention is not instituted at this stage.

DF has been known to occur in both children and adults. However, dengue hemorrhagic fever (DHF) has been traditionally known to be a disease of older children [8]. In recent years, cases of adult DHF have been reported in countries such as Cuba, Puerto Rico, Singapore [9], and Bangladesh. Although older adults account for only a minority of DHF cases, amounting to 2% of DHF cases in Cuba (aged >50 years) [10], 7% in Puerto Rico (aged >60 years) [11], and 3% in Bangladesh (aged >58 years) [12], the prevailing epidemiological trends seem to herald the increasing burden of disease in the elderly. Moreover, with reports of increased mortality in this population [13], there is an urgent need to better understand the management of older adults with dengue.

2. Changing epidemiology

From 1780 to 1940, dengue was transmitted almost exclusively in a pattern of epidemic transmission, where a single virus serotype is introduced into a population of susceptible hosts and mosquitoes, resulting in an explosive outbreak of the disease which subsequently wanes with the decline of susceptible hosts in the population [8]. In this setting, both children and adults in the population are equally affected and the incidence of DHF is usually low [14]. However, the ecological disruption in the Pacific and Southeast Asia following World War II facilitated the transmission of mosquito-borne diseases bringing about an increased frequency of epidemic transmission in these regions, ultimately resulting in a state of hyperendemicity [8].

In a state of hyperendemicity, there is a continuous circulation of multiple dengue virus serotypes in the same area. This requires the year-round presence of competent vector mosquitoes and either a large population or steady movement of individuals into the area to maintain a pool of susceptible individuals, allowing the occurrence of multiple epidemics in a smaller geographic scale. In this setting, the prevalence of antibody against dengue virus rises with age. Thus, children are more likely than adults to experience disease, and most adults in the local population are immune to infection [15]. On a global scale, areas with hyperendemic dengue transmission contribute to the majority of cases of dengue virus infection, thus the burden of disease falls primarily upon the pediatric population.

In recent years, epidemiological studies from regions that traditionally demonstrated a classical pattern of hyperendemic transmission suggested a shift in incidence of disease toward the adult population. In a set of systematic reviews on epidemiological trends of dengue in seven nations, including Brazil, Columbia, the French territories of the Americas, Malaysia, Mexico, the Philippines and Thailand from 2000 to 2011, a trend of increasing adult cases of dengue were noted in five of the seven nations [16–22]. In Brazil, the incidence of dengue virus infections was highest in young adults aged 20–40 years, while in Colombia this was in the group of 15–44 years. In the French territories of the Americas, it was noted that 62% of dengue cases occurred in patients above the age of 15 years. In Malaysia and Thailand, while the majority of cases still occurred in children under the age of 15 years, it was noted by the authors that there has been a gradual but significant shift in cases to individuals aged above 15.

Singapore, ahead of other countries, reported the trend and have put forward several possible explanations for its occurrence. Dengue became hyperendemic in the 1960s, and DHF was a major contributor to childhood deaths. This drove the establishment of an effective vector control program that drastically reduced the Aedes aegypti population from 16% to under 2%, bringing about a 15-year period of low dengue incidence from 1975 to 1990 [23]. When dengue made a resurgence in the 1990s, there was a steady decline in the proportion of patients under 15 years of age and an increase in the proportion of patients above 25 years of age over the years, as well as a trend toward the acquisition of dengue outside the home. A seroprevalence study of adults aged 18–79 years in Singapore comparing the prevalence of anti-dengue IgG antibodies in the population in 2004 and 2010 showed that the age-standardized rates of seroprevalence was significantly lower in 2010 (54 · 4%) compared to 2004 (63 · 1%) [24], implying that a growing proportion of the adult population remained susceptible to dengue virus infections. These findings led the authors to propose that the successful vector control and surveillance program resulted in lowered seroprevalence and herd immunity in the adult population, while the shift in acquisition patterns toward one that was outside of the home (due to better vector elimination in residential versus non-residential sites), resulted in the acquisition of disease at a later stage in life [23].

With the WHO’s push toward an Integrated Vector Management strategy in all its member states in the region of the Americas, Western Pacific and South East Asia, one would expect this shifting epidemiological trend to persist, with the age of disease incidence steadily increasing over time. The once held adage of the elderly population being at low risk for dengue virus infections, due to having acquired immunity from infections earlier in life, may no longer hold true.

3. Clinical presentations and diagnosis in the elderly

The first step in management of dengue and the prevention of dengue-related mortality is early diagnosis and recognition of clinical signs that require urgent interventions. The diagnosis of dengue is heavily dependent on the recognition of clinical signs and symptoms. Both the WHO 1997 and 2009 guidelines on the diagnosis of DF rely on the presence of fever and 2 or more of a list of clinical symptoms, signs or basic laboratory abnormalities (Table 1) [7].

Table 1. Adapted from the World Health Organization dengue: guidelines for diagnosis, treatment, prevention, and control 1997 & 2009.

WHO 1997		WHO 2009
		Travel to or live in dengue endemic area plus
Acute febrile illness with 2 or more of the following:		Fever and 2 of the following:
	Headache	Nausea, vomiting
	Retro-orbital pain	Rash
	Myalgia	Aches and pains
	Arthralgia	Positive tourniquet test
	Rash	Leukopenia
	Hemorrhagic manifestations	Any warning sign
	Leukopenia
And		Warning signs:
	Supportive serology	Abdominal pain or tenderness
Or		Persistent vomiting
	Occurrence at the same location and time as other confirmed cases of dengue	Laboratory: increase in hematocrit concurrent with rapid decrease in platelet count
		Mucosal bleed
		Lethargy, restlessness
		Liver enlargement >2 cm
		Clinical fluid accumulation

In the older adult, the effects of aging on the adaptive and cell-mediated immunity [25] as well as poor cytokine responses to sepsis [26], have been known to result in atypical clinical presentations to various infections. This phenomenon has also been increasingly described with respect to dengue. In a prospective case–control study from Taiwan of 193 adults (31 of whom were aged >65 years old) presenting to the emergency department with serologically confirmed dengue, elderly patients were found to be significantly more likely to present with isolated fever as compared to their younger counterparts (41.9% vs. 17.9%, P = 0.003), and less likely to present with typical symptoms and signs of dengue (41.9% vs. 75.9%, P = < 0.001), which was defined in the study as fever plus at least 1 of following symptoms: bone pain, myalgia, arthralgia, retro-orbital pain, headache, and maculopapular rash [27].

These findings were mirrored in three other studies from Singapore. In a large retrospective study of 6989 adult patients with dengue virus infection, of which 295 were elderly aged 60 and above, it was found that elderly patients were less likely to present with symptoms of headaches (35.3% vs. 49.1%, P = < 0.001), rash (36.6% vs. 47.6%, P = < 0.001), nausea (48.8% vs. 56.1%, P = 0.014), and mucosal bleeding (12.5% vs. 24.2%, P = < 0.001) compared to the younger group [28]. Similar findings were demonstrated in another prospective study from Singapore that looked at 2,129 adult patients who presented with an acute febrile illness to a tertiary hospital. Of these patients, 250 were diagnosed with dengue, and in the subgroup analysis, patients aged above 55 years (n = 43) were found to have significantly less myalgia, arthralgia, retro-orbital pain, and mucosal bleeding [29]. The sensitivity of the WHO 1997 and 2009 diagnostic criteria for probable dengue was only 73.7% (56.9–86.6%, 95% CI) and 81.6% (65.7–92.3%, 95% CI), respectively, in the elderly, as compared to 95.4% (84.2–99.4%, 95% CI) and 100% (91.8–100%, 95% CI) in those aged under 55 years [29]. Rowe et al corroborated similar findings, showing that patients above 60 years of age with laboratory diagnosed dengue were significantly less likely to fulfill the 1997 WHO criteria as compared to those under 60 years of age (93.6% vs. 96.4%, P = 0.014). However, this disparity was not observed with the 2009 WHO criteria [28]. On the other hand, in another retrospective study of 1971 laboratory-confirmed dengue cases, it was found that patients aged above 60 (n = 66) were significantly less likely to report fever (92% vs. 99%, P = 0.002) or have leukopenia (32% vs. 51%, P = 0.002) at the time of presentation [30].

Evidently, the above data suggest that a significant proportion of dengue in the elderly would likely be missed based on current guidelines. The authors suggest that dengue should be considered in the differential diagnosis for elderly patients who reside in or had recent travel to an area-at-risk of dengue who present with an acute undifferentiated febrile illness.

4. Severe dengue and complications

Compared to younger individuals, elderly adults are at increased risk of developing severe dengue and DHF/DSS. This has been consistently reported in studies from various regions in the world, regardless of the state of development of the healthcare system in these areas.

During a dengue outbreak in southern Taiwan in 2002 with 5336 confirmed cases, it was reported that the peak age-specific incidence rate of DHF and DSS occurred in individuals aged 60 years and above, at 70 cases/100,000 persons, as compared to a rate of less than 10 cases/100,000 persons in the younger age groups [13]. Similar findings were reported in a prospective case–control study of 193 adult dengue cases presenting to an emergency department in Tainan in 2007. Patients aged above 65 were more likely to present with DHF as compared to those aged under 65 (12.9% vs. 2.5%, P = 0.02) [27]. Furthermore, in a large epidemiological study from Kaohsiung, Taiwan, spanning from 2003 to 2009 with 2087 laboratory confirmed cases of dengue virus infection, 70% of DHF cases occurred in individuals aged above 55 years [31].

In Singapore, elderly patients aged >60 years had more DHF (29.2% vs. 21.4%, P = 0.002) and severe dengue (20.3% vs. 14.6%, P = 0.006) as compared to younger adults [28]. Similarly, a large cross-sectional study of 6703 laboratory-confirmed cases of dengue in Vitória, Brazil, demonstrated that elderly adults, aged 60–88 years, had an adjusted odds ratio of 1.56 (1.23–1.99, 95% CI) as compared to younger adults, aged 20–59 years, of having severe dengue [32].

Several reasons could have contributed to severe disease in the elderly. First, the process of aging impairs physiological functions [33] and negatively impacts the function of the immune system [26]. In the elderly, monocytes appear to have lower antioxidant response [34] against oxidative stress induced by the dengue virus [35–37]. T cell response and cytokine production in response to dengue virus infection have also been observed to be impaired in the elderly [26].

Second, the probability of acquiring secondary dengue increases with age [38]. In a study from Singapore, the prevalence of secondary dengue in adults aged 60 years and above was almost twice that of those under 60 years (64% vs. 34%, P = 0.0001) [30].

A third factor pertains to the increased prevalence of chronic diseases and other comorbidities in the elderly. In a meta-analysis, the presence of underlying diabetes mellitus was associated with an increased risk for severe clinical presentations of dengue, with an odds ratio of 1.75 (CI: 1.08–2.84, P = 0.022) [39]. A separate systematic review, reported varying multitudes of risk associating severe dengue with preexisting comorbidities; including hypertension, diabetes, allergies, chronic renal disease, previous strokes, chronic obstructive lung disease, and ischemic heart disease. However, as the studies were highly heterogeneous for exposures and outcomes, no definite conclusions could be made. In spite of this, the authors believe that comorbidities are relevant to causing severe clinical dengue and called for further prospective studies to ascertain their effects [40].

Elderly patients with dengue tend to have fewer hemorrhagic manifestations than their younger counterparts [28,32,41]. This difference is most stark in the context of mucosal bleeding, where it was found to be almost twice as common in the younger adults as compared to the elderly (41.1% vs. 21.7%, P = 0.003) [28].

Severe end-organ manifestations of dengue, however, were reported to be more common in the elderly. In a study from Singapore, elderly patients were more likely to have severe organ involvement as compared to younger adults (16.7% vs. 3.1%, P = < 0.001). In particular, liver involvement with hepatomegaly was more commonly seen (3.1% vs. 1.0%, P = 0.006) [28]. Acute renal failure also occurred in higher frequency in individuals aged 65 and above with DHF as compared to those under 65 years of age (12.1% vs. 1.7%, P = 0.001) [33]. However, no significant difference in the prevalence of neurological and cardiac manifestations of dengue has been observed thus far [32].

Conflicting reports have emerged with regards to the occurrence of plasma leakage in the elderly with severe dengue. A study of 17,666 dengue cases from Puerto Rico suggested that the elderly were more likely than youths or younger adults to manifest with plasma leakage [41]. However other studies from Singapore [28,30] and Brazil, did not find any difference in the occurrence of plasma leakage between the two groups [32].

5. Outcomes and mortality

Studies have indicated that when compared with younger adults, the elderly had a longer median length of hospitalization [27,28,33] In addition, it was observed that hospital-acquired infections occurred at greater frequency in the elderly as compared to younger adults (4.9% vs. 1.2%, P < 0.001), with pneumonia and urinary tract infection being the most common [28]. This may be related to the increased length of inpatient stay and higher rates of insertion of medical devices such as indwelling urinary catheters.

Several studies suggested higher mortality in the elderly. There were, however, other conflicting results. Studies from the 2002 dengue outbreak in Taiwan observed that adults aged above 65 had a higher fatality rate as compared to those under 65 (7.6% vs. 0.8%, P = 0.006) [33]. A separate study looking at dengue fatalities during the same outbreak, found that 76% of all patients who died were aged above 55 years [13]. Similarly, in a multicenter retrospective study of adult dengue deaths from 2004 to 2008 in Singapore, the median age of patients who died was 59 years (range 21–86 years) [42]. In the Americas, a retrospective study of dengue cases from 1994 to 1999 in Puerto Rico found that the elderly had a higher risk of death than adults or youths [41]. During the 1981 dengue outbreak in Cuba, a trend toward higher mortality rates in adults aged above 40 years was also observed (1.2 vs. 0.9 deaths per 10,000 dengue virus infections) [43].

On the contrary, Rowe et al. reported that despite higher frequencies of severe dengue and DHF in the elderly, there was no significant difference in the occurrence of admissions to the intensive care unit (ICU) (0.2% in adults vs. 0.7% in elderly, P = 0.13) or death (0.3% in adults vs. 0% in elderly, P = 1.00) [28]. Another study from Taiwan showed no significant difference in the 14-day mortality rate of the elderly (0% vs. 0.6%, P = 1.00) despite higher rates of DHF and a higher Simplified Acute Physiology Score II [44] at the time of presentation [27].

These contrasting mortality outcomes between the studies could partially be accounted for by the differences in socioeconomic status and the availability of healthcare services between the different study populations, given that the data presented here span different countries and time periods. Regrettably, data regarding socioeconomic status and the state of the local healthcare system are largely unavailable across most of these studies, limiting the conclusions that can be drawn from them.

6. Management and prevention of dengue in the elderly

In view of reduced physiological reserves, the effects of immunological senescence, increased frequencies of comorbidities with its associated polypharmacy, and the nuances of the clinical course and outcomes of dengue in this population, we suggest that the clinical management of elderly dengue should be handled differently and with greater caution. Unfortunately, studies in the management of dengue in the elderly are scarce and much of current management strategies are extrapolated from the general population [45].

Judicious fluid therapy is the cornerstone in the management of dengue. Decreased oral intake from poor appetite, increased gastrointestinal losses from vomiting and diarrhea, and plasma leakage can lead to intravascular fluid deficits. The WHO 2009 guidelines recommend giving crystalloids for adults and children with compensated shock, guided by evidence-based fluid management protocols with calculated fluid deficit, weight-based fluid resuscitation over several hours, intensive monitoring of pulse pressure, blood pressure and serum hematocrit, and rapid de-escalation of fluid resuscitation with first signs of clinical improvement care [46]. However, these algorithms were developed based upon studies in the general adult population. Although clinical data are lacking, elderly patients are anticipated to be at a much higher risk of fluid overload from overzealous fluid therapy. This is further exacerbated by the increased risk of acute kidney injury and reduced myocardial function [45]. Noninvasive technologies to assess intravascular fluid status, such as echocardiography or inferior vena cava ultrasound assessment for volume status, should be performed, where available, to direct fluid therapy [47].

Adding to the complexities of fluid management in the elderly, a study found that cardiogenic shock was responsible for 47.1% of all patients who developed shock in a cohort of 162 dengue patients [48]. This suggests that inotropic support instead of fluid replacement may be necessary in a significant proportion of patients with dengue shock to maintain adequate perfusion. Overzealous fluid replacement may instead result in fluid overload and pulmonary edema [46]. Clinical trials on fluid replacement protocols in the elderly are urgently needed.

Although the risk of bleeding manifestations has been shown to occur less frequently in the elderly [28,32,41], a study found that platelet transfusions were carried out more frequently in the elderly population (23% vs. 12%, P = 0.02) [30]. Based on current literature, transfusion of blood products is recommended only in the presence of severe bleeding or clinical suspicion of severe bleeding in a patient with unexplained hypotension [46]. Prophylactic platelet transfusion in adults has been shown to be non-superior to supportive care in preventing bleeding and may instead be associated with adverse events such as anaphylaxis, transfusion-related lung injury, and fluid overload [49]. In light of this, greater emphasis should be placed on the judicious use of blood products, especially in the elderly who already pose a significant challenge in fluid management.

In hospitalized patients, the use of indwelling medical devices such as vascular access and fluid status monitoring must be reviewed regularly and removed when no longer indicated, to reduce the risk of hospital-acquired infections [46].

Preexisting illnesses and polypharmacy in the elderly pose challenges to managing dengue [45]. In a study from South India, 73% of elderly dengue patients (aged 60 years and above) were found to have significant hyponatremia which was attributed to continued diuretic use and poor oral intake during dengue [50]. Antiplatelet agents, vitamin K antagonists, and nonsteroidal anti-inflammatory drugs may increase the propensity for bleeding. An algorithm has been proposed to guide management of anticoagulation therapy in such populations [51]. Antihypertensive agents such as beta-blockers could mask the tachycardia response of compensated dengue shock or exacerbate hypotension [45].

The recent availability of the live attenuated tetravalent chimeric vaccine, Dengvaxia® (CYD-TDV), developed by Sanofi Pasteur, opened avenues for dengue prevention beyond vector-control. Currently it has been licensed in more than 10 countries, including Mexico, Brazil, El Salvador, the Philippines, Costa Rica and Singapore. The phase III studies conducted in Asia and Latin America only included participants from 2 to 16 years old [52,53]. Although the safety data of the vaccine appears robust among adults, there is currently no efficacy studies in adults or the elderly. Further studies into the safety and efficacy of the vaccine in the adult and elderly population are eagerly awaited.

7. Expert commentary

Dengue in the elderly is an increasingly important yet greatly understudied area. Cumulative evidence thus far suggests that clinical manifestations are atypical, making the diagnosis and management of older adults with dengue challenging. There is an urgent need to refine the current diagnostic criteria and identify individuals in this population who are likely to progress to severe disease. Management strategies in this population would have to be adjusted to account for the increased number of comorbidities, and an apparent increase in case fatality. The role of the available dengue vaccines in this population group is uncertain, and more studies into their safety and efficacy are urgently needed.

8. Five-year view

Extrapolating from the current epidemiological trend, we anticipate a progressive shift in the burden of dengue virus infections toward an older population in most regions of the world. This could potentially result in increases in dengue case fatality rates, given the higher risk of mortality in this population. The dearth of clinical trials involving older adults further hinders the implementation of evidence-based management. It is foreseeable that greater demands will be placed on the healthcare system as older adults would require closer monitoring, consequently leading to increased hospitalization rates, longer inpatient stays, and higher demands for intensive care facilities.

Key issues

• Current epidemiological trends have shown a gradual shift in the burden of dengue from the paediatric to the adult population. The incidence of dengue occurring in the elderly has been increasing consequently.

• Dengue in the elderly presents atypically with fewer clinical symptoms and signs. The current WHO diagnostic criteria do not perform as well in this group.

• The elderly are at higher risk of developing severe dengue. This tends to present as severe end-organ involvement instead of haemorrhagic manifestations. Severe plasma leakage may occur more commonly as well.

• Elderly patients with dengue have a higher risk of mortality, are more likely to have a prolonged hospital stay and are at higher risk of hospital-acquired infections.

• Management of the elderly patient with dengue is challenging. Judicious fluid management with accurate assessment of volume status and cardiac function is crucial to avoid causing fluid overload and to guide appropriate initiation of inotropes. Clinicians should be wary of the potential interactions between the elderly patient’s chronic co-morbidities, medications and the clinical course of dengue.

• Efficacy and safety of the recently available dengue vaccine has not been well- established in the elderly population. More studies are urgently needed to delineate its role in this population.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

The manuscript was not funded.