European Journal of Cardiovascular Medicine

Many studies suggest that the current incidence of pleural diseases has increased, and now, the incidence of pleural diseases is around 351–362 per 1,00,000 population, which is close to that of various respiratory diseases with high global burden like bronchial asthma, 270 per 1,00,000, and chronic obstructive pulmonary disease, 260–610 per 1,00,000.[1,2] The reason behind the increase in incidence of pleural diseases is an increase in population of aged people, survivors of malignancies or chronic organ failure. Light’s criteria divide pleural effusion into exudate and transudate effectively. Parapneumonic effusion and empyema are the most common causes of exudative pleural effusion.[3] Large population-based studies have reported high mortality and morbidity in pleural infection among elderly and comorbid patients. RAPID score is a prospectively derived and externally validated risk assessment tool and was developed by Rahman et al.[4] by using multi-centre intra-pleural trial 1 and 2 (MIST1 and MIST2). This score triage pleural infection into three different risk categories, low, medium, and high, based on scores calculated in a patient. RAPID score allows us to predict prognosis in pleural infection without seeing, scanning, or sampling the pleural collection. Patients with parapneumonic effusion usually recovered, but the mortality rate is still around 10% along with long hospital stay, surgical interventions, and high morbidity despite advancement in treatment, especially in empyema.[5,6] In India or any tuberculosis-endemic country, tuberculosis is the most common cause of pleural effusion. RAPID score has not been validated in tubercular patients because most of the studies on RAPID score are retrospective and from non-tubercular endemic countries. This is the first prospective study which directly evaluates the value of RAPID score in both tubercular and non-tubercular patients in the Indian population.

Why do we need this study?

To answer the following questions or filling the gap in knowledge in pleural infection in the Indian prospective.

1.Does RAPID score perform well in tubercular patients as it does in non-tubercular patients, especially in the Indian population?

2.To find out etiological distribution in the Indian population as a cause of pleural infection.

3.Does RAPID score have any role in deciding the management of pleural infection?

Aim

1. Mortality (3-month follow-up) in different risk categories of RAPID score.

2. Length of Hospital stay in different risk categories of RAPID score.
3. Need of surgical intervention in different risk categories of RAPID score.

          Inclusion criteria

1. Study subjects given written informed consent.
2. Age >18 years.
3. Neutrophilic pleural effusion.
4. Bacteriologically positive pleural fluid as per gram stain or culture or CBNAAT.

         Exclusion criteria

1. Study subjects having contraindication for an invasive procedure (e.g., coagulation disorder, thrombocytopenia, etc.).
2. Pleural effusion not suggestive of parapneumonic effusion and empyema
3. Patients having chronic kidney disease.

       Procedure

       This was a tertiary hospital-based prospective type of study. A total 120 study subjects with a parapneumonic effusion and empyema presenting to our in-patient department during the period of 1 year were included. Necessary permission from Ethical Committee and Research Review Board was taken. Patients were admitted in our in-patient department, and these were interviewed thoroughly. Serum urea and serum albumin were sent alongside diagnostic and therapeutic (whenever needed) thoracocentesis, which were performed, and pleural fluid or pus sent for cell count, biochemistry (protein, sugar, etc.), ADA (adenosine deaminase), Gram’s stain and Pyogenic culture, pleural   fluid CBNAAT (cartridge-based nucleic acid amplification test), and sputum CBNAAT were sent for every patient. After compiling all the data, RAPID score was calculated. Patients were divided into risk categories of low (0–2), moderate (3–4), and high (5–7) according to score. The duration of hospital stay (clinically, we assume ≥7 days as longer hospital stays) and the need of surgical interventions were determined, and these were followed for 3 months to look for 3-month mortality in all different risk categories. In the case of the 3-month follow-up term, there were cure, improvement, and deterioration. Those who did not come for follow-up or those who lost to follow-up and did not take appropriate treatment were excluded in final outcome.

       Definitions – At 3-month follow-up to see the patient’s status, we divided them into three main categories: cured, improved, and deterioration. There is no universal definition that can be applied here because both parapneumonic effusion and empyema have different parameters to measure the outcome. For a better understanding of our study, we followed these definitions:

       1.Cured: Parapneumonic effusion and pyogenic empyema: Negative pyogenic culture along with resolution of radiologic evidence of disease and no clinical symptoms at the end of 3-month of follow-up.

       2.Improved: Parapneumonic effusion and pyogenic empyema: Clinical symptoms improved, but there was radiological persistence of disease. In cases of tubercular aetiology, clinical symptoms improved, but there was partial radiological resolution (a decrease in amount of pus/fluid) of disease along with negative follow-up smear at 2 months.

       3.Deterioration: In any patient where there is radiological and clinical disease progression and there is requirement of surgical intervention or the chest drainage tube cannot be removed despite medical therapy at the end of 3-month follow-up.

      Note: In tubercular cases where cure cannot be determined at 3-month as per national tuberculosis elimination programme (NTEP) guidelines, we included them into improved. Patients of parapneumonic effusion and pyogenic empyema: Clinical improvement was decided on the basis of removal of the drainage tube, with no symptoms and no requirement of thoracocentesis. In any case, clinical progression was decided on the basis of persistence of symptoms, inability to remove chest drainage tube, and repeated thoracocentesis requirement.

Rapid score: Rapid score was originally developed by NM Rahman and colleagues in MIST1 trial. Rapid score, which is a prognostic score, includes the following: renal (serum urea), age, purulence, source of infection, and dietary factor (serum albumin).

*Three patients left the study voluntarily as they do not want to participate in the study further.

In our study, a total of 120 patients were included. Among them, the majority of 86 (71.66%) patients had empyema (tubercular and pyogenic), while 34 (28.33%) patients had parapneumonic effusion. As shown in Table 1, Mycobacterium tuberculosis (M. tb) was the most common organism overall, while Pseudomonas was the most common non-tubercular organism and the second overall. Surgery was needed in 17 (14.17%) patients. Its association with RAPID score can be seen in Table 2.

Surgery was needed more in the high-risk category (57.89%) compared to moderate (11.11%), while none in low risk (0%) of RAPID score risk categories. This association was significant (P value < .0001).

Ninety-four patients (78.33%) had long hospital stay (>7 days), while 26 (21.67%) patients had shorter hospital stay (<7) days. Long hospital stay was seen in all patients belonging to moderate- and high-risk categories of RAPID score, while short hospital stay was seen only in the low-risk category. This association was significant as per Table 3 (P value < .0001).

Although Table 4 shows after 3-month follow-up, we divided patients into different categories according to patients’ clinical and radiological statuses. Patients who were declared cured belong to low and moderate risk categories, while none from high risk. Mortality was seen    only in patients in the high-risk category, while none in low and moderate risk.

Association with RAPID score category and intrapleural fibrinolytics use was significant as it shows that intrapleural fibrinolytics use was more in the moderate-risk category as compared to low- and high-risk categories. Intrapleural fibrinolytics was needed in a total of 21 (17.50%) patients out of 120 patients. We used intrapleural fibrinolytics in patients who had adhesions on ultrasonography and empyema/ effusion fully not drained by chest drainage or thoracocentesis.

Previous studies from different geographic areas are compared in Table 5.

Previous studies from different geographic areas are compared.

All these studies are retrospective and done in different geographic areas [Table 5]. All these studies proved prognostic relation of RAPID score in pleural infection. Mortality was seen in the high-risk category of RAPID score, while not in the moderate- or low-risk category of RAPID score. Surgery was needed in all patients in Liou et al.[10] study as this study was the first study to examine the RAPID score and timing of surgery. In this study, they found that late surgery and RAPID score > 3 were independently associated with organ failure and increased mortality. Another four studies have different numbers of patients who required surgery (Simon et al.,[7] Touray et al.,[8] Wong et al.,[9] and Carneiro et al.);[11] similarly, mortalities in different categories of RAPID score were also compared.

Association of RAPID score with all of these (Mortality, need of surgical intervention, Hospital stay, 3-month follow up and intrapleural fibrinolytics) had significant association in non-tubercular patients (P value < .0001), while in tubercular patients, only hospital stay and 3-month follow-up were significantly associated with RAPID score as seen in Table 6. RAPID score predicts prognosis well in both tubercular and non-tubercular patients. It can also predict hospital stay with RAPID score category; long hospital stay was seen in moderate- and high-risk RAPID scores, while short hospital duration was seen only in the low-risk category.

Rewriting of Table 6 was done, and all data are put in n (%).

Among 120 patients, tuberculosis was found in 43 patients, and non-tubercular were 76. In one patient, there was both tuberculosis and pyogenic infection, but it was counted as tubercular in the final outcome as that patient’s sputum CBNAAT was also positive. As in comparison of different studies (Simon et al.,[7] Touray et al.,[8] and Wong et al.)[9] from different geographic areas shows that streptococcus species was the most common organism as a cause of pleural infection.

This study gave us answer to all of our questions which were initially drafted. RAPID score was initially developed as prognostic score in patients of pleural infection and this was also seen in our study. Mortality was only seen in patients having high RAPID score or belonging to high-risk category which is 7 (5.83%) while other studies we compared in Table 6 Touray et al.,[8] Wong et al.,[9] Carneiro et al.,[10] Liou et al.[11] found mortality in 12 (12%), 14 (13%), 5 (14.7%), 11 (6%) respectively and these all patients were from high-risk category of RAPID score. In all patients cause of death was directly related to pleural infection. Cure or improvement were seen less common in high-risk category but more in moderate and low-risk category.[12,13] All the studies mentioned in Table 5 found association between RAPID score and surgery as we did in our study but found different results, except Touray et al.,[8] who found similar results as we did, which was surgery in the high-risk category associated with good results. Simon et al.[7] found surgery in the low- or Patients from the moderate- and high-risk categories will use more healthcare resources and late improvement from disease, which lead to delaying in discharge from the hospital with an increased treatment cost as well because of longer hospital stay, whereas shorter hospital stays are only seen in low-risk category patients, which lead to early discharge (less use of healthcare resources and low cost) and improvement will be more early in this category of patients.

In our study, Mycobacterium tuberculosis was the most common organism, while the second most common and the most common non-tubercular organism was pseudomonas [Table 1]. All other studies compared in Table 6 showed that streptococcus species was the most common organism. This geographical difference in bacteriological pattern suggests that India is a tuberculosis endemic country for causing pleural infection. All the differences were found in comparison of different studies with this study, which might be due to different aetiological agents. Therapeutically, high- and moderate-risk category patients require inpatient care with more aggressive medical management, early chest imaging, and more frequent draining of pus/fluid because of high chances of mortality and further deterioration of patients, while patients from the low-risk category can be managed on an outdoor basis with oral medication and requiring less or single sitting chest drainage of pus/fluid and imaging also. We also try to find out association of intrapleural fibrinolytics with all categories of RAPID score. A total of 21 patients needed intrapleural fibrinolytics, out of which 20 patients were from the moderate-risk category, while only one patient was from the high-risk category and none from the low-risk category.

Strength of RAPID score which is showed in this study are it predict prognosis, tell us timing of surgery and morbidity in pleural infection specially in non-tubercular patients while RAPID score predict prognosis, morbidity in tubercular patients as well. It is given in below risk stratification and comparison. In terms of different aetiologies RAPID score do perform very in all aspects in non-tubercular patients while in tubercular patients it predicts mortality and morbidity. It cannot predict well for surgical interventions in tubercular patients. RAPID score has treatment implications in non- tubercular and complicated tubercular patients.

moderate-risk category had good outcome but the worst in the high-risk category. On the other hand, Carneiro et al.[9] and Wong et al.[11] did not found any association between RAPID score and surgery. Liou et al.[10] and colleagues did the first study regarding timing surgery and outcome. Mortality was increased in patients who had late surgery with the high-risk category of RAPID score. Long hospital stay was noted in all patients of the moderate- and high-risk categories of RAPID score, while short hospital stay was only seen in the

low-risk category. The median hospital stay in our study was 9 days, while hospital stay was longer in John P. Corcoran et al.[14] (13 days) and Sunkaru Touray et al.[8] (17 days).

*Management strictly for non‑tubercular and complicated tubercular patients not for simple tubercular pleural infection patients.

**LOHS – length of hospital stays

The weakness of RAPID score is that it does not have significant association in tubercular patients as compared to non-tubercular patients. It does not predict mortality and surgical intervention in tubercular patients. It also does not predict well in the therapeutic aspect of pleural infections, especially in medical management.

CONCLUSION

RAPID score has good prognostic efficacy, especially in non-tubercular patients, as it helps us to identify the risk category of patients with parapneumonic effusion and empyema. It can also guide clinicians for early management and surgical intervention. Pleural infection needs to be treated early and more cautiously because without appropriate treatment and proper follow-up, it can cause very troublesome complications. This study brought some insight about various aspects of RAPID score, especially in non-tubercular patients, where its association was significant in all parameters, while in tubercular patients, RAPID score performed only in length of hospital stay and prognosis but failed in other parameters. We suggest routine use of RAPID score in patients of pleural infection. This is the first study which used RAPID score in tubercular patients as well and tried to find out some answers.

We propose to encourage further multicentric prospective studies on this topic with a larger study population to further validate and strengthen the prognostic ability of RAPID score to individualise the treatment.

REFERENCES

1. Rycroft CE, Heyes A, Lanza L, Becker K. Epidemiology of chronic obstructive pulmonary disease: A literature review. Int J Chron Obstruct Pulmon Dis 2012;7:457‑94.
2. British Lung Foundation. Asthma Statistics. Available from: https:// statistics.blf.org.uk/asthma. [Last accessed on 2020 Jan 05].
3. Adams FT. Aphorisms by Hippocrates. Available from: Classics.mit.edu/ Hippocrates/aphorisms.html. [Last accessed on 2022 Mar 14].
4. Rahman NM, Kahan BC, Miller RF, Gleeson FV, Nunn AJ, Maskell NA. A clinical score (RAPID) to identify those at risk for poor outcome at presentation in patients with pleural infection. Chest 2014;45:848‑55.
5. Byington CL, Spencer LY, Johnson TA, Pavia AT, Allen D, Mason EO, et al. An epidemiological investigation of a sustained high‑rate pf paediatric parapneumonic empyema: Risk factors and microbiological associations. Clin Infect Dis 2002;34:434‑40.
6. Bradley JS, Byington CL, Shah SS, Alverson B, Carter ER, Harrison C, et al. The management of community acquired pneumonia in infants and children older than 3 months of age: Clinical practice guidelines by the paediatric Infectious Disease Society and the Infectious Diseases Society of America. Clin Infect Dis 2011;53:e25‑76.
7. Simon L, Belaroussi Y, Vayssette A, Khalife T, Le Roux M, Debrosse D, et al. Role of RAPID score and surgery in the management of pleural infection: A single centre retrospective study. J Thorac Dis 2023;15:5340‑8.
8. Touray S, Sood RN, Lindstrom D, Holdorf J, Ahmad S, Knox DB, et al. Risk stratification in patients with complicated parapneumonic effusions and empyema using the RAPID score. Lung 2018;196:623‑9.
9. Wong D, Yap E. Pleural infection in a New Zealand centre: High incidence in Pacific people and RAPID score as a prognostic tool. Intern Med J 2016;46:703‑9.
10. Liou AA, Anderson B, Whitehurst C, Roman S, Beltran C, Acton T, et al. The role of the RAPID score in surgical planning for empyema. J Thorac Dis 2023;15:985‑93.
11. Carneiro DC, Duarte D’Ambrosio P, Mariani AW, Fonini JS, Aguirre GKZ, Carneiro Leão JP, et al. Evaluation of the RAPID score as a predictor of postoperative morbidity and mortality in patients undergoing pulmonary decortication for stage III pleural empyema. Clinics (Sao Paulo) 2024;79:100356.
12. Rahman NM, Maskell NA, West A, Teoh R, Arnold A, Mackinaly C, et al. Intrapleural use of tissue plasminogen activator and DNase in pleural infection. N Engl J Med 2011;365:518‑26.
13. White HD, Henry C, Stock EM, et al. Predicting long‑term outcomes in pleural infections. Ann Am Thorac Soc 2015;12:1310‑6.
14. Corcoran JP, Psallidas I, Gerry S, Piccolo F, Koegelenberg CF, Saba T, et al. Prospective validation of the RAPID clinical risk prediction score in adult patients with pleural infection: The PILOT study. Eur Respir J 2020;56:2000130.