Cross-sectional study investigating the relationship between pit recovery time and serum albumin levels in bilateral lower extremity pitting oedema
===================================================================================================================================================

* Ryosuke Ono
* Ken Horibata

## Abstract

**Objectives** In this study, we re-evaluated the relationship between pit recovery time (PRT) and serum albumin levels and elucidated the factors influencing PRT.

**Design** Cross-sectional study.

**Setting** Patients who visited the outpatient department or were admitted to a small urban hospital in Japan.

**Participants** 135 adult Japanese patients with bilateral lower extremity pitting oedema.

**Interventions** *Primary and secondary outcome measures*: this study assessed the correlation between PRT and serum albumin levels, calculated the predictive accuracy for identifying a group with low albumin levels when the PRT of the lower leg was <40 s, and identified variables that influence PRT.

**Results** We found no significant correlation between lower leg PRT and serum albumin levels. Furthermore, a PRT of <40 s was largely ineffective in predicting low albumin levels. Factors influencing PRT included the diagnosis of malnutrition oedema, examinations conducted during hospitalisation, diagnosis of cardiac oedema, use of diuretics, thickness of the lower limb soft tissue, serum creatinine level, estimated right ventricular systolic pressure (RVSP), age, serum albumin level, potassium level and blood urea nitrogen to serum creatinine ratio. Notable correlations with PRT were observed in relation to lower limb soft tissue thickness, age and estimated RVSP.

**Conclusions** Given that the PRT is influenced by multiple factors, its correlation with serum albumin levels is weak. Thus, predicting hypoalbuminaemia based solely on PRT is inaccurate.

*   INTERNAL MEDICINE
*   GENERAL MEDICINE (see Internal Medicine)
*   Physical Examination

### Strengths and limitations of this study

*   Unilateral leg oedema, leg oedema with significant differences between the right and left sides and non-pitting oedema were excluded.

*   All the measurements were taken by the same examiner.

*   To identify factors that could influence pit recovery time (PRT), we comprehensively collected information that could affect oedema, including medications.

*   Our measurement method is less objective than that used by Henry and Altmann who employed a specialised instrument and Vernier scale to ensure uniformity in their measurements.

*   In a single-centre study conducted in a rural small hospital in Japan, the majority of the study population consisted of elderly individuals, and the presence of concurrent acute illnesses may have influenced PRT and serum albumin levels.

## Introduction

Lower extremity oedema is a common symptom in primary care settings. Differential diagnoses of oedema are numerous,1 2 and the initial differentiation is performed using information such as the timing of onset of oedema, whether it is bilateral or unilateral, medication history, presence of systemic diseases and presence of pitting oedema.1 2

Henry and Altmannroposed a method for differentiating pitting oedema in the lower extremities by considering the time required for a manually induced pit to recover (pit recovery time, PRT).3 They suggested a division at 40 s: if the PRT is less than 40 s, it is classified as fast oedema, and hypoalbuminaemia is suspected; if the PRT is 40 s or more, it is considered slow oedema, and other causes are more likely. This method is being used in clinical practice. In their study, they used a special tool to create a pit in the oedema above the tibia for 31 patients who exhibited lower extremity pitting oedema within the past 3 months. The PRT was measured until the pits disappeared. The results indicated a correlation between the logarithm of the PRT and serum albumin levels (p<−0.001). From the plotted diagram, a PRT cut-off of 40 s was established.

However, our clinical experience frequently demonstrates that rapid oedema does not necessarily correlate with hypoalbuminaemia. Moreover, several aspects of the research by Henry and Altmann3 warrant further investigation, such as the reasoning behind using 40 s as the cut-off value, the lack of explicit statement on the predictive accuracy including sensitivity and specificity, the presence of outliers in the plot diagram without discussion of their causes and the unidentified factors that might influence PRT. Although this study was performed in 1978, no subsequent research investigating the PRT was found. Therefore, in our study, we aimed to re-evaluate the relationship between PRT and serum albumin levels in pitting oedema of the lower extremities, to determine the accuracy of predicting hypoalbuminaemia from fast oedema, and to investigate factors other than serum albumin that might influence PRT.

## Methods

### Study design

This was a cross-sectional study in a single hospital.

### Study population

This study included patients aged 15 years or older (to exclude children) who visited the outpatient department or were admitted to the Department of Internal Medicine at the Kameyama Municipal Medical Centre in Japan from 1 May 2021 to 31 March 2023. We recruited patients who exhibited pitting oedema distal to the knee joint. We excluded patients with oedema in unilateral lower limb, significant asymmetry in oedema between the left and right sides, non-pitting oedema, oedema due to localised skin inflammation or injury (such as cellulitis, burns and anaphylaxis), undergoing dialysis and individuals with critical life-threatening conditions who did not have the time for PRT measurement or examination. Based on a prior study by Henry and Altmann, we estimated that our sample size should be more than 126 cases. The calculation method was as follows: first, we inferred data from the plot diagram of their study and categorised it into two groups, a low albumin group with serum albumin levels less than 3.5 g/dL and a normal albumin group with serum albumin levels of 3.5 g/dL or more. We calculated the average PRT scores for each group. The results were ~53 and 117 s, respectively, with an average difference in PRT of approximately 64 s between the two groups. The SD for both groups was about 55 s, yielding an effect size of 1.0 or more. As the prior study had a small number of cases, we estimated the effect size to be small at 0.5 and calculated the sample size using an alpha value (two-sided) of 0.05 and a beta value of 0.20, resulting in a requirement of more than 63 cases per group.

### Data collection

Data collected from the medical interview included age, sex, admission status (inpatient/outpatient), body height, body weight, Birthel Index, Charlson Comorbidity Index, medication, medical history, paralysis, snoring, onset of oedema and exacerbation of oedema depending on body position or time of day. The variables related to the physical examination included, examined in a sitting or lying posture, skin changes related to lower limb varicose veins,4 soft tissue thickness in the oedema and PRT. Laboratory variables included haematocrit, albumin, total protein (TP), sodium (Na), potassium (K), blood urea nitrogen (BUN), serum creatinine (sCr), BUN/sCr, glucose, estimated plasma osmolarity (Posm), C-reactive protein (CRP) and brain natriuretic peptide (BNP). The variable being assessed in the transthoracic echocardiography were inferior vena cava (IVC) diameter, respiratory variation of the IVC, estimated transtricuspid pressure gradient and estimated right ventricular systolic pressure (RVSP).5

### Methods for measuring PRT and soft tissue thickness in the oedema

Lower leg oedema is measured at the midpoint on the rough surface of the tibia at the distal 1/3 height of the line connecting the medial malleolus and medial condyle of the tibia,3 and foot oedema is measured at the dorsum of the third metatarsal head, which is the midpoint of the line connecting the heads of the first and fifth metatarsals.6 For each measurement of soft tissue thickness, the distance from the skin surface to the anterior surface of the tibia was measured for lower leg oedema, and the distance from the skin surface to the head of the third metatarsal was measured for foot oedema, using an ultrasound device. The examiner then firmly pressed the oedema with the thumb for 2 s until a clear indentation remained and subjectively measured the number of seconds (up to a maximum of 600 s) it took for the indentation to completely disappear. This was performed by the same examiner for all patients. If indentations from socks or compression stockings are observed, they were removed and measurements taken after the indentations have disappeared. The average of the soft tissue thickness at all locations is defined as the ‘lower limb soft tissue thickness’. The average of the PRT at the two lower leg locations is defined as the ‘lower leg PRT’, the average PRT at the two feet locations is defined as the ‘foot PRT’, and the average PRT at all locations is defined as the ‘lower limb PRT’.

### Diagnosis of the cause of oedema

Two general practitioners skilled in diagnosing oedema will make judgements based on medical chart data, algorithms for the differential diagnosis of bilateral lower-limb oedema1 2 7 and reviews of each oedema pathology.8–10 Please note that the PRT was not used as a reference. Additionally, if necessary for the diagnosis of lower limb oedema, appropriate additional tests (such as urinary protein concentration, lower limb venous ultrasound and abdominal imaging) will be conducted.

### Statistical analyses

Data compilation was performed using Microsoft Excel, and data analysis was conducted using EZR on R-commander (V.1.41 for Windows).11 Continuous variables were expressed as means and SD. The correlation between two variables was evaluated using Pearson’s product-moment correlation coefficient, whereas the correlation between binary variables was assessed using Pearson’s χ2 test. If over 20% of the cells had expected counts of less than 5, Fisher’s exact test was used. The imputation of missing data was conducted through available case analysis.

1.  Correlation between PRT and serum albumin levels: for both lower leg and foot PRT, the correlation with serum albumin levels was plotted and the correlation coefficient of the two variables was calculated.

2.  Calculation of the predictive accuracy for identifying a low albumin group when the lower leg PRT is less than 40 s: as proposed by Henry and Altmann,3 the lower leg PRT is dichotomised into the ‘fast oedema group’ (less than the cut-off value of 40 s) and the ‘slow oedema group’ (equal to or greater than the cut-off value). The serum albumin level is also dichotomised into a ‘low albumin group’ (below the cut-off values of 2.0, 2.5, 3.0, 3.5 g/dL, respectively) and a ‘normal albumin group’ (equal to or above the cut-off value). Using the rapid oedema group as test-positive and the low albumin group as disease-positive, a 2×2 contingency table was created, and the sensitivity, specificity, positive likelihood ratio and negative likelihood ratio for predicting the low albumin group from a lower leg PRT of less than 40 s were calculated.

3.  Selection of variables that influence PRT: Survival analysis was conducted with the lower limb PRT serving as the survival time and the disappearance of pitting as the event occurrence. For the binary survey items, a log-rank test was performed using these binary variables as the grouping variables, and items with a p-value less than 0.2 are selected. The selected binary and continuous survey items were utilised as explanatory variables in a Cox proportional hazards analysis performed using the backward selection method, with items yielding a p-value less than 0.05. For the chosen continuous variables, the correlation between each of these variables and the PRT was plotted, and the correlation coefficient between the two variables was calculated.

### Patient and public involvement

Patients and the public are not being involved in the development of the research questions and outcome measures, the study’s design, the assessment of interventions in the study or recruitment and conduct. We will disseminate the final results to the study participants after these are published in a peer-reviewed journal.

## Results

### Baseline characteristics

During the study period, 162 patients developed lower limb oedema. Of these, 20 were excluded because they met the exclusion criteria. An additional seven patients who declined to participate in the study were also excluded, resulting in 135 participants (figure 1). In total, pitting oedema was confirmed at 473 sites across the lower legs and feet of 135 participants (~3.5 sites per person). Of these, 12 participants had oedema at both lower leg sites, 14 participants had oedema at both foot sites, six participants had oedema at both lower leg sites and one-foot site, nine participants had oedema at one lower leg site and both foot sites and 94 participants had oedema at all four sites. The participants’ background information is provided in table 1. Missing data were not present in most of the data, including PRT and albumin values. However, height and weight had missing values in 41 cases, and the onset of oedema was unknown for 21 cases as patients did not recall. Regarding the cause of oedema, 100 cases had a single cause, 33 had two causes and two cases had three causes. The primary causes were cardiac oedema (60 cases), drug-induced oedema (26 cases), malnutrition-related oedema (18 cases), rheumatoid-related oedema (9 cases), hepatic oedema (5 cases), nephrotic syndrome-related oedema (4 cases), venous insufficiency-related oedema (3 cases), lipid oedema (2 cases), inferior vena cava syndrome-related oedema (2 cases) and undetermined causes (6 cases).

![Figure 1](http://bmjopen.bmj.com/https://bmjopen.bmj.com/content/bmjopen/14/1/e079327/F1.medium.gif)

[Figure 1](http://bmjopen.bmj.com/content/14/1/e079327/F1)

Figure 1 
Flow chart of participant inclusions, measurements and statistical analysis.

View this table:
[Table 1](http://bmjopen.bmj.com/content/14/1/e079327/T1)

Table 1 
Characteristics of participants and test results

### Correlation between PRT and serum albumin levels

No significant correlation was found between lower-leg PRT and serum albumin levels (p=0.26, correlation coefficient 0.10 (95% CI: −0.08 to −0.28))(figure 2), and no correlation was observed between foot PRT and serum albumin levels (p=0.55, correlation coefficient 0.06 (95% CI: −0.12 to 0.23)).

![Figure 2](http://bmjopen.bmj.com/https://bmjopen.bmj.com/content/bmjopen/14/1/e079327/F2.medium.gif)

[Figure 2](http://bmjopen.bmj.com/content/14/1/e079327/F2)

Figure 2 
Correlation between serum albumin levels and oedema in the lower legs and dorsal feet.

### Calculation of the predictive accuracy for identifying a low albumin group when the lower leg PRT is less than 40 s

Regardless of the value used for serum albumin, sensitivity ranged from 0.36 to 0.5, specificity ranged from 0.59 to 0.61 (table 2) and the likelihood ratio ranged from 0.83 to 1.26. Therefore, rapid oedema is largely unable to predict low albumin levels.

View this table:
[Table 2](http://bmjopen.bmj.com/content/14/1/e079327/T2)

Table 2 
Accuracy of predicting low albumin levels based on pit recovery time of less than 40 s in lower leg oedema.

### Selection of variables that influence PRT

Binary variables with p<0.2 in the log-rank test included inpatient (p<0.001), diagnosis of malnutrition oedema (p<0.001), use of diuretics (p=0.01), use of limaproalfadex (p=0.01), diagnosis of cardiac oedema (p=0.03), history of liver disease (0.05), use of β-blocker (p=0.06), diagnosis of oedema due to nephrotic syndrome (p=0.1) and paralysis (p=0.17). The binary variables that were significant in the COX proportional hazards analysis included the diagnosis of malnutrition oedema (p=0.001), inpatient (p=0.03), diagnosis of cardiac oedema (p=0.03) and use of diuretics (p=0.03). The significant continuous variables in the COX proportional hazards analysis were lower limb soft tissue thickness (p<0.001), sCr level (p<0.001), estimated RVSP (p=0.002), age (p=0.002), serum albumin level (p=0.01), potassium level (p=0.02) and BUN/sCr ratio (p=0.03) (table 3). Furthermore, significant correlations were found between lower limb soft tissue thickness (p<0.001), age (p=0.01) and estimated RVSP (p=0.02) among the selected continuous variables. The respective correlation coefficients were −0.33 (95% CI: −0.48 to −0.17), 0.22 (95% CI: 0.05 to 0.37) and 0.20 (95% CI: 0.03 to 0.36).

View this table:
[Table 3](http://bmjopen.bmj.com/content/14/1/e079327/T3)

Table 3 
Variables influencing PRT obtained using the COX proportional hazards model with backward stepwise selection

## Discussion

In this study, we investigated the relationship between PRT and serum albumin levels in patients presenting with pitting lower limb oedema. However, we were unable to confirm the relationship between these two variables reported by Henry and Altmann,3 and found it challenging to predict hypoalbuminaemia using a cut-off of 40 s for the lower leg PRT. Nevertheless, the serum albumin level was identified as a variable influencing PRT, along with several other variables.

A possible reason for our inability to adequately predict serum albumin levels from PRT could be that much like oedema formation is multifactorial, PRT may also be determined by multiple factors; thus, the impact of serum albumin levels on PRT may be minimal. According to the traditional Starling model, oedema formation is largely influenced by differences in capillary hydrostatic and osmotic pressures.12 It was presumed that the manifestation of oedema in hypoalbuminaemia was due to a single mechanism, where a decrease in colloid osmotic pressure (COP) reduces the albumin gradient from the intravascular to the interstitial compartment, increasing fluid movement from the intravascular to the interstitial compartments. However, there has been accumulating evidence suggesting that the Starling model is unable to accurately predict fluid behaviour, leading to a proposal for a revision of this principle by Levick and Michel.13 Reports have also suggested that factors other than a decrease in intravascular COP contribute to changes in fluid distribution in hypoalbuminaemia caused by nephrotic syndrome,14 and that pathologies other than hypoalbuminaemia are involved in the formation of oedema in malnourished children.15 Thus, oedema formation during hypoalbuminaemia is believed to involve a complex mechanism involving multiple factors. Currently, it is unknown whether PRT is strongly affected by a single factor or a combination of multiple factors. However, this study revealed that many variables influence the PRT, implying that the PRT is determined by multiple factors. Therefore, it is difficult to accurately predict albumin levels, one of the causes of PRT, using the PRT, which is formed by the complex influence of multiple factors.

In contrast, in this study, the serum albumin level was identified as a variable influencing PRT. It seems plausible, as reported by Henry and Altmann,3 that serum albumin levels indeed have an impact on PRT. Notably, the diagnosis of malnutrition-related oedema was also selected as a variable that influenced PRT, and it had a higher HR than serum albumin level. This may be because patients diagnosed with malnutrition-related oedema are likely to have other diseases excluded, and the influence of variables other than serum albumin levels on PRT may be minimal. Therefore, it may be possible to predict serum albumin levels from PRT in situations where the influence of other variables has been excluded, and it may also be possible to predict from PRT that the cause of oedema is malnutrition related. Further studies are required to determine the diagnostic accuracy of PRT.

Numerous variables influencing PRT have been identified, some of which have not been previously examined and warrant further investigation. First, regarding the soft tissue thickness that was observed to have a negative correlation with PRT, soft tissue thickness is primarily an indicator of the degree of oedema.16 Therefore, the results of this study suggest that the more severe the oedema, the shorter the PRT. Previous studies have reported a correlation between the visual extent of oedema and hypoalbuminaemia,15 suggesting a high likelihood of relationships among soft tissue thickness, PRT and serum albumin levels. However, as soft tissue thickness can be determined by the causes of oedema, similar to PRT, attention must be paid to the confounding factors that can influence both PRT and soft tissue thickness. For example, fat deposition and increased subcutaneous connective tissue could be potential causes.10 17 Second, several variables related to cardiac oedema were identified and a trend toward prolonged PRT for cardiac oedema was confirmed. This can be understood by considering that cardiac oedema occurs through capillary hydrostatic pressure differences mediated by the estimated RVSP. The estimated RVSP, which had a positive correlation with PRT, is an indicator of cardiac oedema,18 and it is suggested that as the RVSP increases, capillary hydrostatic pressure rises and oedema forms.19 Additionally, cardiac oedema is susceptible to renal dysfunction due to cardiorenal interactions. The use of diuretics and creatinine levels are also believed to be related to cardiac oedema. The diagnosis of cardiac oedema was selected as a variable with a HR>1. In a separate analysis, a univariate analysis using a COX proportional hazard analysis with only the diagnosis of cardiac oedema as an explanatory variable resulted in a HR<1.0 and a significant difference, suggesting that it was corrected by other variables in the multivariate analysis. However, it is unclear which variables are influenced and how they are influenced. Third, as age increased, there was a trend of PRT becoming shorter, and inpatients tended to have a shorter PRT. These variables are susceptible to the effects of underlying diseases and nutritional status. This suggests that PRT can vary with differences in the underlying diseases due to patient background and setting, indicating the limitations of PRT in terms of diagnostic accuracy.

Our study has several limitations. First, we evaluated the PRT by subjectively measuring the time taken for pitting to disappear, and all the measurements were taken by the same examiner. However, this methodology may lack the reproducibility of the approach used by Henry and Altmann,3 who employed a specialised instrument and Vernier scale to ensure uniformity in their measurements. A significant distinction is that, while Henry and Altmann created a 5 mm pitting using a special tool, our study saw variances in pitting depth, as it was induced solely by thumb pressure. This variation suggests that our measurement method is less objective than that used by Henry and Altmann, which may account for our inability to demonstrate a correlation between the two variables. Nonetheless, it is important to consider the practical limitations in a clinical setting, where employing a special instrument to induce pitting, as in Henry and Altmann’s method, is not feasible. Furthermore, there is inherent variability in the pitting depth created by different examiners. Hence, the diagnostic accuracy of PRT using our measurement method might have a higher degree of applicability in clinical practice. Second, our study participants were individuals who visited or were admitted to hospitals equipped with inpatient beds. This implies a greater probability of including individuals with acute diseases that necessitate hospitalisation, as opposed to the population typically observed for oedema. Consequently, the presence of these simultaneous acute diseases might have influenced PRT and serum albumin levels. Third, only 31% of PRT measurements were taken in a sitting position, which may introduce variations in PRT due to gravity’s influence.

## Conclusions

We further investigated the relationship between PRT and serum albumin levels in patients with pitting lower limb oedema. However, we found no significant correlation between the PRT and serum albumin levels. Predicting hypoalbuminaemia using the method proposed by Henry and Altmann,3 which sets a cut-off of 40 s for lower leg PRT, proved to be challenging. We believe that PRT is determined by multiple variables, including soft tissue thickness in the oedema and estimated RVSP, and not just serum albumin levels. Thus, it is difficult to accurately predict serum albumin levels using PRT alone.

## Data availability statement

Data are available in a public, open access repository. Extra data can be accessed via the Dryad data repository at [http://datadryad.org/](http://datadryad.org/) with the doi: 10.5061/dryad.6wwpzgn59.

## Ethics statements

### Patient consent for publication

Consent obtained directly from patient(s).

### Ethics approval

This study involves human participants and was approved by Kameyama Municipal Medical Center Medical Research and Ethics Committee (code No.2021042701-2). Participants gave informed consent to participate in the study before taking part.

## Acknowledgments

The authors express their gratitude to MT and MK, internal medicine physicians at Kameyama Municipal Medical Center, for referring patients for this study.

## Footnotes

*   Contributors Author contributions: RO and HK conceived and designed the study. RO, HK, and MT recruited patients for the study. RO collected data. RO and HK analysed and interpreted the data. RO wrote the manuscript. All authors have approved the final manuscript. RO is responsible for the overall content as guarantor.

*   Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

*   Competing interests None declared.

*   Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

*   Provenance and peer review Not commissioned; externally peer reviewed.

[http://creativecommons.org/licenses/by-nc/4.0/](http://creativecommons.org/licenses/by-nc/4.0/)
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: [http://creativecommons.org/licenses/by-nc/4.0/](http://creativecommons.org/licenses/by-nc/4.0/).

## References

1.  Trayes KP, Studdiford JS, Pickle S, et al. Edema: diagnosis and management. Am Fam Physician 2013;88:102–10.
    
    

2.  Ely JW, Osheroff JA, Chambliss ML, et al. Approach to leg edema of unclear etiology. J Am Board Fam Med 2006;19:148–60. [doi:10.3122/jabfm.19.2.148](http://dx.doi.org/10.3122/jabfm.19.2.148)
    
    [Abstract/FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NToiamFiZnAiO3M6NToicmVzaWQiO3M6ODoiMTkvMi8xNDgiO3M6NDoiYXRvbSI7czoyNjoiL2Jtam9wZW4vMTQvMS9lMDc5MzI3LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

3.  Henry JA, Altmann P. Assessment of Hypoproteinaemic edema: A simple physical sign. Br Med J 1978;1:890–1. [doi:10.1136/bmj.1.6117.890-a](http://dx.doi.org/10.1136/bmj.1.6117.890-a)
    
    [FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6MzoiUERGIjtzOjExOiJqb3VybmFsQ29kZSI7czozOiJibWoiO3M6NToicmVzaWQiO3M6MTA6IjEvNjExNy84OTAiO3M6NDoiYXRvbSI7czoyNjoiL2Jtam9wZW4vMTQvMS9lMDc5MzI3LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

4.  Lurie F, Passman M, Meisner M, et al. The 2020 update of the CEAP classification system and reporting standards. J Vasc Surg Venous Lymphat Disord 2020;8:342–52. [doi:10.1016/j.jvsv.2019.12.075](http://dx.doi.org/10.1016/j.jvsv.2019.12.075)
    
    

5.  Rudski LG, Lai WW, Afilalo J, et al. Guidelines for the echocardiographic assessment of the right heart in adults: A report from the American society of echocardiography. Journal of the American Society of Echocardiography 2010;23:685–713. [doi:10.1016/j.echo.2010.05.010](http://dx.doi.org/10.1016/j.echo.2010.05.010)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1016/j.echo.2010.05.010&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=20620859&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 
    
    [Web of Science](http://bmjopen.bmj.com/lookup/external-ref?access_num=000279067900002&link_type=ISI) 

6.  Kogo H, Higashi T, Murata J. Reliability of a new practical evaluation method for Pitting edema based on the depth of the surface imprint. J Phys Ther Sci 2015;27:1735–8. [doi:10.1589/jpts.27.1735](http://dx.doi.org/10.1589/jpts.27.1735)
    
    

7.  Patel H, Skok C, DeMarco A. Peripheral edema: evaluation and management in primary care. Am Fam Physician 2022;106:557–64.
    
    

8.  Mahmood SB, Nelson E, Padniewski J, et al. Polymyalgia Rheumatica: an updated review. Cleve Clin J Med 2020;87:549–56. [doi:10.3949/ccjm.87a.20008](http://dx.doi.org/10.3949/ccjm.87a.20008)
    
    [Abstract/FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NToiY2Nqb20iO3M6NToicmVzaWQiO3M6ODoiODcvOS81NDkiO3M6NDoiYXRvbSI7czoyNjoiL2Jtam9wZW4vMTQvMS9lMDc5MzI3LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

9.  Klein-Weigel PF, Elitok S, Ruttloff A, et al. Inferior vena cava-syndrome. Vasa 2021;50:250–64. [doi:10.1024/0301-1526/a000919](http://dx.doi.org/10.1024/0301-1526/a000919)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1024/0301-1526/a000919&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=33459041&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 

10. Canning C, Bartholomew JR. Lipedema. Vasc Med 2018;23:88–90. [doi:10.1177/1358863X17739698](http://dx.doi.org/10.1177/1358863X17739698)
    
    

11. Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical Statistics. Bone Marrow Transplant 2013;48:452–8. [doi:10.1038/bmt.2012.244](http://dx.doi.org/10.1038/bmt.2012.244)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1038/bmt.2012.244&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=23208313&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 
    
    [Web of Science](http://bmjopen.bmj.com/lookup/external-ref?access_num=000316920100021&link_type=ISI) 

12. Starling EH. On the absorption of fluids from the connective tissue spaces. J Physiol 1896;19:312–26. [doi:10.1113/jphysiol.1896.sp000596](http://dx.doi.org/10.1113/jphysiol.1896.sp000596)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1113/jphysiol.1896.sp000596&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=16992325&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 

13. Levick JR, Michel CC. Microvascular fluid exchange and the revised starling principle. Cardiovasc Res 2010;87:198–210. [doi:10.1093/cvr/cvq062](http://dx.doi.org/10.1093/cvr/cvq062)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1093/cvr/cvq062&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=20200043&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 

14. Bhave G, Neilson EG. Body fluid Dynamics: back to the future. J Am Soc Nephrol 2011;22:2166–81. [doi:10.1681/ASN.2011080865](http://dx.doi.org/10.1681/ASN.2011080865)
    
    [Abstract/FREE Full Text](http://bmjopen.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiam5lcGhyb2wiO3M6NToicmVzaWQiO3M6MTA6IjIyLzEyLzIxNjYiO3M6NDoiYXRvbSI7czoyNjoiL2Jtam9wZW4vMTQvMS9lMDc5MzI3LmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

15. Gonzales GB, Njunge JM, Gichuki BM, et al. The role of albumin and the extracellular matrix on the pathophysiology of edema formation in severe malnutrition. EBioMedicine 2022;79:103991. [doi:10.1016/j.ebiom.2022.103991](http://dx.doi.org/10.1016/j.ebiom.2022.103991)
    
    

16. Kogo H, Murata J, Murata S, et al. Validity of a new quantitative evaluation method that uses the depth of the surface imprint as an indicator for Pitting edema. PLOS ONE 2017;12:e0170810. [doi:10.1371/journal.pone.0170810](http://dx.doi.org/10.1371/journal.pone.0170810)
    
    

17. Schellong SM, Wollina U, Unger L, et al. Leg swelling. Internist (Berl) 2013;54:1294–303. [doi:10.1007/s00108-013-3339-z](http://dx.doi.org/10.1007/s00108-013-3339-z)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1007/s00108-013-3339-z&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=24264570&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 

18. Shah MG, Cho S, Atwood JE, et al. Peripheral edema due to heart disease: diagnosis and outcome. Clin Cardiol 2006;29:31–5. [doi:10.1002/clc.4960290108](http://dx.doi.org/10.1002/clc.4960290108)
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=16477775&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 

19. Cho S, Atwood JE. Peripheral edema. Am J Med 2002;113:580–6. [doi:10.1016/s0002-9343(02)01322-0](http://dx.doi.org/10.1016/s0002-9343(02)01322-0)
    
    [CrossRef](http://bmjopen.bmj.com/lookup/external-ref?access_num=10.1016/S0002-9343(02)01322-0&link_type=DOI) 
    
    [PubMed](http://bmjopen.bmj.com/lookup/external-ref?access_num=12459405&link_type=MED&atom=%2Fbmjopen%2F14%2F1%2Fe079327.atom) 
    
    [Web of Science](http://bmjopen.bmj.com/lookup/external-ref?access_num=000179453900008&link_type=ISI)