Delayed diagnosis and treatment of deep vein thrombosis - an underrecognized factor for its related outcomes?

Abstract

Background

See More

Rapid diagnosis and treatment of deep vein thrombosis (DVT) reduce morbidity and mortality.

Objective

This study aimed to investigate the time to diagnostic imaging, time to treatment, and outcomes associated with the delayed treatment of DVT.

Methods

We retrospectively investigated 110 patients who were diagnosed with proximal DVT in outpatient clinics in a single academic centre in Thailand from January 2019 to September 2020. We recorded demographic data, clinical presentations, time to diagnostic imaging (ultrasound of the legs), and time to treatment. We recorded outcomes, including death, pulmonary embolism (PE), recurrent DVT, and thrombus resolution 3 months after the diagnosis of DVT.

Results

Of all 110 patients, 42 (38.2%) were male. Median age (IQR) was 68.5 (58–80) years. The median (IQR) time to diagnostic imaging was 7 days (2–27). The median (IQR) interval from the first OPD visit to the initiation of treatment was 14 days (3–31). The delayed diagnosis (more than 7 days from the first clinic visit to diagnostic imaging) was observed in 54 (49%) patients with DVT. The delay in treatment (more than 7 days from the first clinic visit to treatment) was observed in 67 (60.9%) patients. The maximum time to treatment was 160 days. Death and PE occurred in 9% and 33% of patients with delayed treatment, while they occurred in 5% and 20% of those with early treatment, respectively. No recurrent DVT was observed. Among 25 patients who had follow-up imaging, residual thrombus occurred in 71% of patients with delayed treatment and 63% of those with early treatment.

Conclusion

Half of the DVT patients had delays in diagnosis and treatment of proximal DVT. The routine request option was associated with a delayed diagnosis. The importance of early diagnosis and treatment of DVT should be raised among physicians, and improvement strategies are warranted.

Introduction

Venous thromboembolism (VTE), which includes deep vein thrombosis (DVT) and pulmonary embolism (PE), is a relatively common and potentially life-threatening condition. The annual incidence of first-time VTE is approximately 100 per 100,000 individuals [1]. Among patients with symptomatic proximal DVT without chest symptoms, up to 50% may develop PE if the condition is left undiagnosed or untreated. Without treatment, 50% of symptomatic VTEs have recurrent thrombosis within 3 months [2]. Furthermore, mortality within seven days of VTE diagnosis has been reported to be as high as 25% in unselected patient populations [3]. In the case of DVT, delayed diagnosis and treatment have also been identified as risk factors for the development of post-thrombotic syndrome (PTS), a chronic complication that significantly impairs quality of life [4, 5].

In Asian populations, the incidence of VTE is generally lower compared to Western countries [6]. Common etiologies include malignancy, hospitalization, and post-surgery [7]. Among hereditary thrombophilia, protein S and protein C deficiency are more frequent, while factor V Leiden and prothrombin gene mutation are rare in Asia [8].

Given that anticoagulant therapy is highly effective in reducing thrombus progression and recurrence, early diagnosis and prompt initiation of treatment are essential to reduce the risk of fatal and recurrent VTE. Therefore, timely recognition and management of VTE are critical to minimizing both morbidity and mortality associated with this condition [9].

Despite the availability of effective anticoagulant therapy, delays in the diagnosis and treatment of VTE remain common, with reported rates ranging from 20 to 50% depending on the definitions used [10,11,12]. Studies indicated that multiple and severe clinical symptoms, rather than patient VTE risk profile were associated with earlier diagnosis of VTE [8]. Other factors contributing to diagnostic delay include limited patient awareness, atypical or mild symptom presentation, geographic barriers, and healthcare system constraints such as limited imaging availability and scheduling inefficiencies. Provider-related issues, including under recognition of VTE in low-risk or atypical cases, also contribute. Understanding and addressing these barriers is critical to improving timely diagnosis and reducing VTE-related morbidity and mortality.

In current clinical practice, we observe that many patients continue to experience delays in diagnosis and treatment. Therefore, we conducted this study to assess the time to diagnosis and treatment of proximal DVT, determine the proportion of patients with delayed diagnosis and treatment, and identify factors associated with these delays.

Materials and methods

Study design

We conducted a retrospective study at a single-center academic public hospital in Thailand. The objectives were to determine the duration from the first outpatient clinic visit to the diagnosis of symptomatic proximal DVT, confirmed by imaging, to determine the duration from the first outpatient clinic visit to the treatment of symptomatic proximal DVT. The secondary objectives were to assess the proportion of patients with delayed diagnostic imaging and delayed treatment and to study factors associated with delayed diagnosis and complications of DVT occurring within three months of diagnosis. The study was approved by the Ethical Research Committee of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University (Approval No. MURA2021/465).

A list of patients who underwent Doppler ultrasound of the lower extremities between January 2019 and September 2020 was retrieved from the hospital database. Patients diagnosed with proximal DVT were identified. It is noted that in our institution, the standard protocol is to perform whole-leg compression ultrasound, regardless of whether the request is for unilateral or bilateral imaging. The inclusion criteria were an age of more than 18 years, symptomatic DVT, and evaluation in an outpatient setting. We did not include inpatients in the study because, in our hospital, the ultrasound scheduling system prioritizes inpatients, resulting in shorter imaging waiting times. Additionally, inpatients often have comorbidities that could influence treatment outcomes. Exclusion criteria included patients undergoing preoperative DVT screening and those who had ultrasounds performed for follow-up of a prior DVT.

Data collection

We retrospectively reviewed electronic charts of patients diagnosed with proximal DVT and collected relevant clinical and demographic data. Demographic variables included age, sex, and underlying conditions such as hypertension, diabetes, dyslipidemia, cancer, chronic kidney disease, or other comorbidities. Patient status, including whether they were bedridden, was recorded. Information regarding the department responsible for patient care was also collected, including surgery, medicine, rehabilitation, palliative care, radiology, gynecology, home visit services, orthopedics, and family medicine. The type of clinic where the patient was initially seen was categorized as either a general clinic or a premium clinic. A premium clinic was defined as a specialized private outpatient clinic within our institution. The physician type was categorized as either resident/fellow or staff. Details of the ultrasound request were recorded, including the urgency level (routine, urgent, or stat) and whether the request was for a single or bilateral lower extremity evaluation. The urgency of the request was determined at the discretion of the clinician ordering the imaging. The stat option required the physician to contact and discuss the case with a radiologist by phone before confirming the order. The clinical presentation using Well’s criteria for DVT and the probability of DVT was assessed [13]., if available. This scoring system classifies patients into three probability categories: low probability (−2 to 0 points), moderate probability (1 to 2 points), and high probability (3 to 8 points).

We defined the duration of key time intervals as follows: the time to DVT diagnosis was the interval between the date of the outpatient clinic visit when the clinician ordered the test and the date the diagnostic imaging was performed. A delay in diagnosis was defined as an interval of more than seven days, whereas an early diagnosis was defined as an interval of seven days or less. The time to treatment was defined as the total interval from the date of the outpatient clinic visit, when the test was ordered, to the date anticoagulant therapy was initiated. A delay in treatment was defined as an interval of more than seven days from the first outpatient clinic visit to treatment initiation. The time from diagnostic imaging to treatment was defined as the interval between the date the imaging was performed, and the date anticoagulant therapy was initiated.

Complications of DVT, including death, pulmonary embolism, recurrent DVT, post-thrombotic syndrome, and thrombus resolution at three months after diagnosis, were recorded.

Statistical analysis

Baseline characteristics were reported by descriptive analysis. Mean with standard deviation (SD) or median with interquartile range (IQR) were reported as appropriate. The difference in frequencies between the two groups was analyzed using a Chi-square test. Mann-Whitney U test was used to test continuous variables between 2 groups, while the Kruskal-Wallis H test was used to test among those with more than 2 groups. A p-value of < 0.05 was considered statistically significant. Univariate and multivariate logistic regression models were performed to determine factors associated with delayed timing of diagnosis imaging (> 7 days). All statistical analyses were performed using STATA version 17.0.

Results

A total of 110 patients with newly diagnosed proximal DVT were included in the study. The median age (IQR) of patients was 68.5 (58–80) years. 42 (38%) patients were male and 68 (62%) were female. 21% of patients were in bedridden status. Common underlying diseases included hypertension (52%) and malignancy (45%). Based on the Wells score, common clinical presentations were pitting edema (55%), entire leg swelling (46%), and difference in calf diameter > 3 cm compared with asymptomatic leg (31%). According to the Well’s score, 52% and 48% had moderate and high probability of DVT, respectively. None of them had a low probability of DVT. The majority of DVTs were managed by physicians in the Department of Medicine (69%), Department of Surgery (17%), and others in 14%. Other baseline characteristics are shown in Table 1.

Table 1 Characteristics of 110 DVT patients included in the study

The median (IQR) time from the first outpatient clinic visit to diagnostic imaging was 7 days (2–27). The median (IQR) interval from the first outpatient visit to the initiation of treatment was 14 days (3–31). The median (IQR) time from diagnostic imaging to treatment was 0 days (0–5). The delayed diagnosis (more than 7 days from the first outpatient visit to diagnostic imaging) was observed in 54 (49%) patients with DVT. The delay in treatment (more than 7 days from the first outpatient clinic visit to treatment) was observed in 67 (60.9%) patients. The maximum time to treatment was 160 days. Table 2 summarizes the key intervals, and Table 3 shows the duration from the first clinic visit to diagnostic imaging. Figure 1 shows the distribution of time from the first clinic visit to diagnostic imaging and treatment.

Table 2 Timing intervals

Table 3 Duration from the first clinic visit to diagnostic imaging

Fig. 1

Distribution of time intervals from outpatient (OPD) visit to diagnostic imaging and treatment. Boxplots represent timing intervals in days. The green boxplot shows the interval from the initial outpatient clinic visit to the day diagnostic imaging was performed. The blue boxplot shows the interval from diagnostic imaging to treatment initiation. The gray boxplot represents the total time from outpatient clinic visit to treatment initiation and is the sum of the green and blue intervals for each patient

As for treatment, all patients received anticoagulant therapy. 52% received the treatment on the day of the ultrasound reports. This was primarily because radiologists directly notified the physicians upon detecting abnormal findings. No patient had empirical therapy with anticoagulants before the confirmed diagnosis.

In the univariate analysis, both physician type (resident/fellow vs. staff) and request option (routine vs. urgent/stat) showed associations with a p-value less than 0.15, warranting their inclusion in the multivariable model. Following multivariable analysis, the request option was a significant independent factor influencing diagnostic delay. Specifically, when an imaging request was designated as routine (adjusted OR 3.00, 95% CI 1.35–6.70; P = 0.01) compared to cases where an urgent/stat request was made. Being under the care of resident/fellow physicians showed a trend towards an increased likelihood of delayed diagnosis (adjusted OR 2.25, 95% CI 0.98–5.15; P = 0.06) when compared to staff physicians (Table 4).

Table 4 Factors associated with delayed timing of diagnosis imaging (> 7days) in univariate and multivariable analyses. Odds ratios (OR) and adjusted odds ratios (aOR) with 95% confidence intervals (CI) are reported. Reference indicates the baseline comparison category for categorical variables in regression analysis. Premium clinic refers to patients evaluated in a specialized private outpatient clinic

Conversely, characteristics such as the ordering physician’s Department (e.g., surgery, others vs. medicine), whether the patient was seen in a premium clinic, the site of ultrasound (single vs. bilateral), and the various components of the clinical presentation based on Wells’ criteria (e.g., active cancer, entire leg swelling, previously documented DVT, etc.) did not demonstrate a significant association with delayed diagnosis in either the univariate or multivariable models.

Outcomes are shown in Table 5. Death occurred in 5% of the early treatment group and 9% of the delayed group. Among 13 follow-up scans for PE, 3 cases were diagnosed. PE was found in 20% of the early group and 33% of the delayed group. No recurrent DVT occurred in either group. Among 25 follow-up scans, residual thrombus was detected in 63% of the early group and 71% of the delayed group. There were no significant differences in the complications between the two groups. Data on PTS was not documented in most patients’ medical records. Therefore, this outcome was not reported.

Table 5 Complications related to DVT at 3 months in patients with early (≤ 7 days) and delayed (> 7 days) treatment of DVT

Discussion

In this study, we found that about half of patients with symptomatic proximal DVT experienced delays in both diagnosis and treatment. The median time from the first outpatient clinic visit to diagnostic imaging was 7 days and the median time to treatment initiation was 14 days. A routine request was the factor associated with a delayed diagnostic timeline. Complications, including death, PE, and residual thrombus, were numerically higher in the delayed diagnosis group but did not differ significantly between groups.

A prospective study in North America conducted in early 2000 demonstrated a substantial delay in the diagnosis of DVT and PE in approximately 20% of patients [12]. The definition of delay diagnosis in this study was more than 7 days. Reasons for a delayed diagnosis include delays in seeking medical evaluation rather than the time from evaluation to confirmatory testing. Other factors that may contribute to delays in seeking medical attention and diagnosis include lower educational status, retirement, geographic distance from healthcare facilities, and comorbidities such as age-related or neurodegenerative conditions. These factors can affect symptom recognition, decision-making, and access to care.

In contrast, our study demonstrates the imaging wait time is the most time-consuming process. Although we did not have data on symptoms onset to the first clinic visit, it implied that the delay in diagnosis and treatment would be a more concerning issue than the previous studies. Interestingly, the maximum time to VTE diagnosis and treatment was 160 days in symptomatic proximal DVT patients. This is an unexpectedly long duration and could contribute to consequent morbidity.

An observational study in Italy demonstrated that delayed diagnosis (more than 10 days from onset of symptoms) was observed in 22.6% of patients with VTE. In DVT patients, factors associated with earlier diagnosis were the presence of multiple signs or symptoms, the presence of pain, and previous venous thrombosis [11]. However, in our study, intermediate or high-risk clinical probability of DVT did not confer to early diagnosis. Still, only the request option (on urgency level) of the confirmed imaging by the physician did. It is noted that both studies in North America and Italy included both patients with PE and DVT.

A study in Czech Republic demonstrated a delay treatment > 7 days in more than half of the patients with DVT. They also demonstrated a correlation between delayed treatment and the rate of incomplete recanalization of proximal and distal thrombosis [14]. A study of DVT patients in Iran demonstrated the mean interval from symptoms onset to initiation of treatment was 4.7 days. The interval from the first visit to confirmation of diagnosis was 2 days (SD 4). In this study, patients with an intermediate and high probability of DVT had significantly earlier diagnoses of DVT [10]. All studies mentioned demonstrated a delay process from symptom onset to the date of first medical evaluation, which suggests that improving patients’ awareness about DVT and PE would be a strategy for improvement.

Our study demonstrates that delays in diagnosis and treatment may not be solely related to patient factors but are also influenced by healthcare-related factors. We demonstrate that the urgency level assigned to the imaging request plays a critical role in timely diagnosis. Urgency or stat request option required physician to contact radiologist by phone, it might be burdensome for some physicians, potentially contributing to less frequent use of this pathway. Acknowledging this issue raises awareness and encourages the development of strategies for quality improvement. Clinical probability should be incorporated into clinical practice to identify patients at higher risk for DVT. Training bedside ultrasound for DVT screening would be useful because if the physicians can perform the screening ultrasound by themselves, the treatment might be started earlier while waiting for confirmation from the radiologists.

The reimbursement system may influence the timeliness of VTE diagnosis and management, particularly in private hospital settings. However, our hospital is a public institution where all patients are covered under the universal healthcare scheme; therefore, all costs related to diagnostic imaging and treatment are fully reimbursed. However, our hospital operates both general and “premium” clinics. Premium clinics utilize a separate appointment system and may involve out-of-pocket payment, though civil servants receive partial reimbursement. However, our analysis revealed no significant association between clinic type (general vs. premium) and delays in DVT diagnosis or treatment initiation.

The strength of our study is that it is the first study in Thailand to demonstrate the time to treatment and diagnosis, the proportion of patients with delayed diagnosis and treatment, and the resulting complications. Data were manually reviewed retrospectively, and all DVT cases were confirmed through imaging studies. However, we acknowledge several limitations. This was a single-center, retrospective study, which may limit the generalizability of our findings, particularly to settings with different healthcare contexts, such as private hospitals. The small sample size also restricted our ability to compare outcomes between the two groups due to insufficient statistical power. Additionally, we did not collect data on patients who underwent Doppler ultrasound but were not diagnosed with DVT. As a result, we could not assess whether the number of Doppler ultrasound requests was excessive or contributed to an increased workload for radiologists. There was also missing data, particularly regarding PTS outcomes, preventing us from reporting on this aspect. Furthermore, we lacked data on the interval between symptom onset and the first medical evaluation due to incomplete records, and we did not collect data on educational status, geographic distance from healthcare facilities, and comorbidities such as age-related or neurodegenerative conditions that could affect to healthcare access.

Conclusion

Half of the DVT patients had delays in diagnosis and treatment of proximal DVT. The routine request option was associated with a delayed diagnosis. The importance of early diagnosis and treatment of DVT should be raised among physicians, and improvement strategies are warranted.