Development of rapid evaluation scale for pericardiectomy of tuberculous constrictive pericarditis by 12-lead electrocardiogram

Abstract

Aim

We developed a rapid evaluation scale for pericardiectomy through a 12-lead electrocardiogram (ECG), in order to improve the diagnostic accuracy of pericardiectomy of tuberculous constrictive pericarditis.

Method

In this study, 262 patients with tuberculous constrictive pericarditis (102 patients) and non-tuberculous constrictive pericarditis (160 patients) were selected by convenience sampling method as participants in Hangzhou Red Corss Hospital from January 2018 to April 2023. The expert validity analysis was carried out by cross-sectional investigation combined with the results of the previous expert questionnaire to establish 12-lead ECG-based the rapid evaluate scale for pericardiectomy of tuberculous constrictive pericarditis. Cronbach’α coefficient and Spearman-Brown coefficient were used for the internal consistency reliability of the scale. Exploratory factor analysis was used for structural validity. Receiver operating characteristic (ROC) was used to analyze the sensitivity and specificity of this scale.

Results

After preliminary verification, the cronbach’α coefficient and Spearman-Brown coefficient were 0.830 and 0.800, respectively. Three factors with feature roots greater than 1.0 were extracted by principal component analysis and maximum variance orthogonal rotation, and the cumulative contribution rate was 57.77%. Each loading factor ranges from 0.427 to 0.863. The sensitivity of the scale was 96.1%, the specificity was 96.9% and the Yorden Index was 93.0%.

Conclusions

The 12-lead ECG-based rapid evaluation scale for pericardiectomy of tuberculous constrictive pericarditis has high sensitivity, specificity and accuracy, and has certain scientific research and clinical application value.

Introduction

Tuberculous constrictive pericarditis (TCP) is a chronic inflammatory process that involves the pericardial wall and visceral layer, leading to pericardial thickening, limiting diastolic activity and thereby reducing heart function [1]. TCP is characterized by thickening and calcification of pericardial fibers, which impairs diastolic filling, reduces cardiac output, and ultimately leads to heart failure [2]. The clinical symptoms of TCP are atypical and include progressive dyspnea, fatigue, multiple plasma effusion, edema, cough, weakness, and palpitations, which leads to poor early diagnosis of chronic TCP [3]. Endotracheal intubation and general anesthesia through the median sternal incision pericarditis is usually the first choice for the clinical treatment of TCP [4]. At present, it is generally believed that timing of surgery is an important factor affecting the TCP prognosis, and early surgery may be associated with a better prognosis [5]. Given the severe complications associated with untreated TCP, timely surgical intervention, such as pericardiectomy, is critical. The important role of pericardiectomy is to remove the fibrous thickened pericardium, restore diastolic filling and increase cardiac output [6]. Venous pressure tubes were routinely placed to monitor central venous pressure. Changes in central venous pressure before and after surgery were used as the pericardial release effect, and central venous pressure ≤ 14 cm H₂O was used as the reference standard for pericardiectomy.

An electrocardiogram (ECG) is the most common test that records the electrical activity of the heart produced during each heart cycle [7]. ECG parameters can reflect cardiac dysfunction such as arrhythmia, myocardial hypertrophy, myocardial infarction. In addition, Morris et al. introduced for the first time the V1 lead P-wave terminal force (Ptfv1) measured by a 12-lead electrocardiogram, which can reflect the severity of heart disease [8]. However, few studies at home and abroad have found that 12-lead electrocardiogram, a non-invasive, non-radiative, convenient and efficient examination method, can be used as an important means of clinical evaluation of TCP pericardiectomy [9]. Therefore, this study developed a rapid evaluation scale for TCP pericardiectomy (RESTP), which provided an important auxiliary diagnostic tool.

Compared with the previous determination of indications for pericardiectomy, this study together included Ptfv1 value, sinus tachycardia, ST segment changes and other related ECG results. Furthermore, this study innovatively calculated the weight of indicators, and designed the scoring rules. This study aims to develop a rapid evaluation scale using 12-lead ECG parameters to provide clinicians with a reliable tool for determining the need for pericardiectomy in TCP patients, potentially improving early diagnosis and treatment outcome.

Data and methods

Sample collection

In this study, convenience sampling method was used to select 262 patients with TCP (102 patients) and non-TCP (160 patients) from Hangzhou Red Cross Hospital as subjects of this study. Among them, the patients with non-TCP included healthy people. Clinical data such as personal basic data, physical examination data and electrocardiogram data were collected. The developed RESTP was used to evaluate whether pericardiectomy needed to be performed. The objective is to use this scale to quickly quantify the conditions for pericardiectomy based on existing medical influence means, so that patients can obtain surgical treatment more effectively, timely and accurately.

Inclusion and exclusion criteria

Patients with TCP should be confirmed according to tuberculin test, T. SPOT-TB test, CT and echocardiography, and microbiology. Microbiological confirmation mainly includes smear and culture confirmation of pericardial effusion. Patients with non-TCP (including healthy people) were not clinically diagnosed with TCP. All subjects should have no other organ diseases, infectious diseases, blood diseases, anesthesia allergies, surgical contraindications, and malignant tumors. All subjects should have informed consent. This study has been approved by the Medical Ethics Committee of Hangzhou Red Cross Hospital (Approval number: [2023] Review No. 125).

Endotracheal intubation with general anesthesia pericardiectomy through median sternal incision

The patient was placed in a supine position with the chest protruding from the shoulder blade area of the back and splitting the median sternum. If there is retrosternal adhesion, separate the adhesion and use a thoracotomy to stretch the sternum on both sides. The pericardium is removed from the apex of the heart. In addition, pericardial adhesion is light, pericardial thickening is not obvious, and it is easy to peel. Cut the thickened pericardium with a blade one by one. There is often a loose connective tissue between the thickened pericardium and the outer membrane, in order to properly peel off the interface of the pericardium. After opening the thickened pericardium, the beating heart can be seen protruding outwards. After separating part of the pericardium, the assistant gently lifts the pericardium with pliers, and the surgeon gently presses the surface of the heart with his left hand to fully reveal the degree of adhesion between the thickened pericardium and the myocardium. If the adhesion is loose, it can be bluntly separated with finger gauze or peanut forceps, and the force part of separation should be on the pericardial surface. In case of cord or ribbon adhesion, sharp separation with scissors or surgical blades is required. If the adhesion is very tight, the original separation site should be abandoned, and the pericardium should be re-cut and separated at other sites, that is, easy first and difficult later, to make separation easier and minimize the damage to the heart muscle and the heart and blood vessels. The scope of dissection was determined according to the patient’s cardiac function and the degree of pericardial adhesion. The basic scope of the general stripping: should be the apex of the heart to be completely stripped; The left side is close to the left phrenic nerve; The fibrous constriction ring of atrioventricular groove and inferior vena cava entrance must be loosened. The sequence of dissection should be left ventricular → right ventricular outflow tract → atrioventricular coarctation ring → inferior vena cava ring band. For other details, we referred to this Mohamed Al-Kazaz’s practices for pericardiectomy [10].

ECG measurements

All patients underwent a standard 12-lead resting ECG in the supine position. The 12-lead ECG data was recorded by a trained ECG physician at double amplitude (20 mm/mV) and paper speed (50 mm/s). ECG parameters include heart rate, time from the start of the QRS complex to the end of the T wave (QTc) after correction, ST segment change, T wave change and so on, where Ptfv1 was measured and calculated as follows: (1) Five continuous P waves were selected and measured in sinus rhythm; (2) When the p wave is in the vertical state, the Ptfv1 value is denoted as zero; (3) When the positive and negative ends of P-wave deflection occur, mark the intersection point of the horizontal line derived from the P-wave starting point and the P-wave descending line, and then record the vertical diameter and horizontal distance between the P-wave intersection point and the end point as the amplitude and duration of the negative end of P-wave deflection, respectively; (4) The average amplitude (mm) and duration (s) of the negative deflection at the end of the P-wave are calculated according to 5 consecutive amplitudes and duration; (5) The product of average amplitude (mm) and duration (s) is denoted as Ptfv1 (mm•s) value. ECG parameters such as Ptfv1 value, heart rate, bundle branch block or arrhythmia were recorded.

Score calculation of RESTP

In this study, the developed RESTP was used to evaluate whether pericardiectomy needed to be performed. The objective is to use this scale to quickly quantify the conditions for pericardiectomy based on existing medical influence means, so that patients can obtain surgical treatment more effectively, timely and accurately.

The indexes and scoring rules of RESTP: (1) Ptfv1 value ≤ -0.04 mm•s (+ 3 points), -0.04 mm•s ≤ Ptfv1 value ≤ -0.036 mm•s (+ 1 point); (2) Sinus tachycardia (+ 1 point); (3) ST segment change (+ 1 point); (4) T wave change (+ 1 point); (5) Limb lead low voltage (+ 1 point); (6) Low voltage of chest lead (+ 1 point); (7) QRS time limit of limb lead ≤ 60 ms (+ 2 points); (8) QRS time limit of limb lead > 80 ms (-2 points); (9) Abnormal Q wave (-2 points); (10) Left ventricular high voltage (-2 points). Items 1–7 were set as bonus points and items 8–10 as subtractive points of the rapid Diagnosis scale.

Statistical analysis

In this study, the data of the survey objects were input into EXCEL software and SPSS23.0 statistical software for analysis, and the general basic information of the survey objects was statistically described. Quantitative variables were described statistically using Mean (Mean) and standard deviation (SD), while qualitative variables were described statistically using frequency and percentage. Cronbach’α coefficient and Spearman-Brown coefficient were used for internal consistency reliability test. Exploratory factor analysis was used for structural validity. The sensitivity and specificity of the RESTP were analyzed by using the receiver working curve.

Results

Demographic characteristics

A total of 102 patients with TCP and 160 patients with non-TCP were investigated. The mean age of patients with pericarditis was 57.87 years (SD = 15.23) and that of patients without pericarditis was 58.21 years (SD = 13.40). There were more males in pericarditis group than females, and the proportion was higher than that in non-pericarditis group. The subjects in the pericarditis group were mostly farmers and workers (92.20%), while those in the non-pericarditis group were only 13.20%. Most of the students in the pericarditis group had primary school education or junior high school education (92.20%), while those in the non-pericarditis group had high school education or above (90.10%). The marital status of the two groups of respondents is mostly married, both of which are more than 90.00%. The number of subjects in the pericarditis group (38.20%) was higher than that in the non-pericarditis group (12.50%). The proportion of hypertension in the non-pericarditis group (40.60%) was higher than that in the pericarditis group (18.60%). The number of people with diabetes in the non-pericarditis group (25.60%) was higher than that in the pericarditis group (12.70%). The number of patients with coronary heart disease in the non-pericarditis group (6.30%) was higher than that in the pericarditis group (0%). There were no cardiomyopathy and congenital heart disease in both groups. See Table 1 for details.

Table 1 Demographic information of the subjects

Reliability and validity test

The reliability and validity of the developed RESTP were analyzed in clinical application, which were used in patients with TCP and non-TCP. It’s reliability and validity were calculated to verify its practicability and effectiveness.

Reliability of RESTP

Reliability refers to the consistency or stability of measurement results. Internal consistency is the most commonly used criterion. In this study, cronbach’α coefficient and Spearman-Brown coefficient reflecting internal consistency were used to test the internal consistency of the rating scale. The results showed that the cronbach’α coefficient and the Spearman-Brown coefficient were 0.830 and 0.800, respectively. The internal consistency reliability of the scale was good.

Validity of RESTP

Structural validity

Structural validity can reflect the degree of consistency between the structure of the rapid diagnosis scale and the theoretical framework. This study intends to use exploratory factor analysis to extract common factors from all variables. If the variation that can be explained by common factors is greater than 50%, and the load of each item on the corresponding common factors is greater than or equal to 0.4, then the structural validity is good.

The results showed that the KMO statistic of this scale was 0.827. The approximate chi-square value of Bartlett spherical test is 795.329 (p < 0.001), statistically significant, suitable for factor analysis. Three factors with feature roots greater than 1.0 were extracted by principal component analysis and maximum variance orthogonal rotation method, and the cumulative contribution rate was 57.77%, as shown in Table 2. The load factor of each index ranges from 0.427 to 0.863, as shown in Table 3.

Table 2 Explanation of total variation of the rapid diagnostic scale for pericardial dissection

Table 3 Factor load matrix after rotation

Sensitivity and specificity analysis

ROC curve was used to analyze the sensitivity and specificity of the RESTP, and the optimal threshold was determined to evaluate the prediction of pericardiectomy in patients with TCP. The optimal threshold was determined according to the Yorden index. The ROC curve of this scale was shown in Fig. 1. The area under the curve (AUC value) was 0.991, and the 95% confidence interval was (0.950, 0.995). The sensitivity and specificity values of each threshold point on the ROC curve are shown in Table 4. The optimal threshold was 3.5, and the sensitivity (sensitivity), specificity (specificity) and Yorden index of the rapid diagnosis scale for pericardiectomy were 0.961, 0.969 and 0.930, respectively, and the positive likelihood ratio and negative likelihood ratio were 31.00 and 0.040, respectively.

Fig. 1

ROC Curve

Table 4 Sensitivity and specificity values on the ROC curve

The area under ROC curve (AUC value) can reflect the screening efficiency of this scale, and the closer it is to 1, the more accurate it will be. It is generally considered to have moderate diagnostic value when it reaches 0.8–0.9. The AUC of the RESTP in this study is 0.991, which has high diagnostic value. Sensitivity of this scale was evaluated through identifying probability of patients who really need pericardiectomy. Specificity is used to measure the ability of this scale for identifying patients who do not need pericardiectomy. In this study, the sensitivity and specificity of the optimal threshold were 96.1% and 96.9%, indicating that the RESTP has good sensitivity and specificity. However, sensitivity and specificity cannot fully reflect the authenticity of this scale, so this study included the Yorden index and likelihood ratio. The Yorden index, also known as the correct diagnosis index, is the sum of sensitivity and specificity minus one. The closer to 1, the better the screening efficiency, indicating that this scale for pericardiectomy can correctly judge whether pericardiectomy is necessary. The positive likelihood ratio indicates how many times of the patients with the “need to perform” result is compared with the “need not perform” for this scale. The higher the ratio, the higher the screening value of the scale. The negative likelihood ratio indicates how many times the probability of “no pericardiectomy required” results in patients with pericardiectomy is higher than that in patients with “no pericardiectomy required”. The smaller the ratio, the higher the screening value of this scale. In this study, the optimal threshold of Yorden index, positive likelihood ratio and negative likelihood ratio were 0.930, 31.00 and 0.040, respectively, indicating that the evaluation value of this scale for pericardiectomy was high.

Discussion

Tuberculous constrictive pericarditis is a chronic inflammatory process that involves the pericardial wall and visceral layer, leading to pericardial thickening, limiting diastolic activity and thereby reducing heart function. It is characterized by thickening and calcification of pericardial fibers, impairs diastolic filling, reduces cardiac output, and ultimately leads to heart failure. Pericardiectomy is an effective way to correct hemodynamic abnormality and improve cardiac function in constrictive pericarditis [11]. Once constrictive pericarditis is diagnosed, surgery should be performed as early as possible, and the earlier the operation, the better the prognosis [12]. Clinical symptoms, chest X-ray, CT, echocardiography and other auxiliary examinations as well as routine placement of venous pressure monitoring tubes are often used as surgical indications. But each has its pros and cons.

First, diagnosis is made according to clinical symptoms. Constriction of the pericardium restricts the diastolic expansion of the ventricles, thus reducing stroke volume and cardiac output. Reduced cardiac output causes the kidneys to retain water and sodium, resulting in increased blood volume. Constriction of the pericardium makes it difficult for venous blood to return to the heart. Therefore, there are clinical manifestations of elevated venous pressure, such as jugular vein irritation, liver enlargement, ascites, abdominal fluid, lower limb edema and other signs. The left heart is compressed by the scar, which may cause pulmonary circulation congestion and dyspnea. Second, according to chest X-ray, CT and echocardiography, echocardiography indicated diffuse or localized pericardial thickening, atrial enlargement, ventricular contraction or cardiac dysfunction. Chest X-ray and CT examination revealed pericardium thickening and calcification. The third routine is to place a venous pressure tube to monitor the central venous pressure. The change of central venous pressure before and after surgery is used as the pericardial release effect, and the central venous pressure ≤ 14 cm H₂O is used as the reference standard for dissection. The clinical manifestations are increased venous pressure, such as jugular vein irritability, liver enlargement, ascites, abdominal fluid, lower limb edema and other signs, not only the symptoms of constrictive pericarditis, but almost all diseases leading to right heart failure will appear this clinical symptom. The evaluation of constrictive pericarditis based on chest X-ray, CT and echocardiography, although more intuitive has great advantages, but there are many drawbacks. Chest X-ray film, CT in addition to the long inspection time often need to make an appointment and charge a high fee, the most important radiation damage to the human body, and the inspection will be affected by some interference factors such as chest gas accuracy, when the bag exudate is thick, there is often a certain difficulty in identifying pericardial constriction. The cost of echocardiography is not low, the accurate instrument is not convenient to move, the bedside instrument lacks accuracy, and the examination is affected by some interference such as sternum and ribs and the sampling section, and there will be false negative display of pericardial constriction. As a diagnostic reference, central venous pressure is a static indicator that indirectly reflects cardiac preload through pressure volumetric method. Measured by the upper and lower vena cava or the internal tube of the right atrium, central venous pressure is one of the main indicators for clinical observation of hemodynamics. It is affected by three factors: right heart pumping function, circulating blood volume and vascular tension of systemic venous system. Increased pressure in the superior vena cava system contributes to the early diagnosis of constrictive pericarditis. Some studies have suggested that the initial CVP value varies greatly among patients, and the initial CVP value alone cannot be used to predict and evaluate volume responsiveness well. CVP refers to the pressure in the right atrium of the upper and lower vena cava patients in the thoracic cavity, and is an indicator reflecting the preload of the right atrium. However, CVP is a static indicator that indirectly reflects cardiac preload through pressure volumetric method, and factors such as receptor site, surgical procedure, ventricular compliance, vascular tension, and thoracic pressure [13, 14].

Studies at home and abroad rarely mention the use of electrocardiogram, a non-invasive, non-radiative, low cost, convenient and efficient examination, as a reference evaluation of TCP pericardiectomy. Therefore, this study developed a rapid evaluation scale based on 12-lead ECG to evaluate pericardiectomy in patients with TCP, which provides a certain reference basis for clinical surgery of TCP.

Scientific analysis of RESTP

Excellent reliability

Reliability means that a survey is repeated on the same group of objects and the consistency trend of the results is observed. Retest reliability, duplicate reliability and internal consistency reliability are common reliability measurement indexes [15]. At present, the widely used evaluation methods of internal consistency reliability are broken half reliability method and cronbach’α coefficient method. The former is to divide the item into two halves and then calculate the correlation coefficient of the scores of the two parts, so as to avoid the influence of the survey objects by time changes. And cronbach’α coefficient method can avoid the shortcomings of split half reliability method. Therefore, in this study, the reliability of the RESTP was evaluated by the method of split half reliability and cronbach’α coefficient. As early as 1965, some scholars proposed a measurement standard based on cronbach coefficient: α coefficient < 0.35, low reliability; 0.35–0.80 is medium confidence; α coefficient > 0.80, high reliability. In this study, the Spearman-Brown coefficient and cronbach’α coefficient were 0.800 and 0.830 respectively. It can be seen that the reliability of this scale is high.

Excellent validity

In this study, exploratory factor analysis was used to test the structural validity of the RESTP. The evaluation of factor analysis was mainly based on two points: (1) the common factor was consistent with the assumed structure of the system design and the cumulative contribution rate was at least 40%; (2) the load value of each item on its common factor was > 0.4. The cumulative contribution rate of the three common factors extracted in this study is as high as 57.77%. The factor load values of all the 10 indicators are > 0.4. On the whole, this scale has excellent structural validity.

Excellent sensitivity and specificity

The pericardium has a protective effect on the heart, preventing the ventricles from overexpanding to maintain a constant blood volume [16, 17]. Patients with chronic constrictive pericarditis have reduced cardiac output and reduced stroke volume. For patients with more severe constrictive pericarditis or those who are physically active, the heart rate increases to maintain cardiac output, which can lead to more serious complications [18]. Pericardial exfoliation may result in different outcomes for different patients. Therefore, the rapid diagnostic scale developed in this study integrated Ptfv1 value, sinus tachycardia, ST segment change, T wave change, limb lead low voltage, thoracic lead low voltage, limb lead QRS duration ≤ 0.06 mm, limb lead QRS duration > 80ms, abnormal Q wave and high voltage of left ventricle were used to evaluate the surgical indications. The area under ROC curve of the scale was 0.991, and the optimal threshold was 3.50. At this time, the sensitivity, specificity and Yorden index of the scale were 0.961, 0.969 and 0.930, respectively. Patients identified by the scale as “needing pericardial dissection” were 96% likely to need it according to the doctor’s diagnostic criteria. Therefore, this RESTP has high accuracy, strong practicability, non-invasive, non-radiative, convenient and efficient advantages in judging whether patients have pericardiectomy surgery. It can be used as an important auxiliary diagnostic tool for clinicians to evaluate pericardiectomy surgery, so as to make up for the shortcomings of the current surgical indication rapid evaluation tools.

Limitation

This study has several limitations. This scale was tested for content validity, expert validity, and construct validity, but did not undergo confirmatory construct testing. Future studies should focus on further validity tests of structural equation modelling. All respondents were in one medical center in Zhejiang. While reliability and validity were shown for the instrument, generalizability could only come after administering the rapid diagnostic scale to other groups in different locales and confirming the reproducibility of findings.

Conclusion

The 12-lead ECG-based RESTP of tuberculous constrictive pericarditis has excellent sensitivity, specificity and accuracy. In addition, the low cost, easy detection, non-invasive and non-radiative characteristics of ECG make RESTP unique compared to existing methods. Overall, RESTP ensures the accuracy and has the advantages of easy to use and rapid diagnosis. The scale has certain scientific research and clinical application value. It can be used as an important auxiliary diagnostic tool for clinicians to evaluate pericardial dissection surgery, so as to make up for the shortcomings of the current surgical indication rapid evaluation tools. This study represents a significant step forward in the diagnostic evaluation of TCP, providing a non-invasive, reliable, and accessible tool that has the potential to improve patient outcomes through timely surgical intervention.