Acute Exacerbation of Idiopathic Pulmonary Fibrosis

Abstract

Episodes of Acute exacerbation (AE) of Idiopathic Pulmonary fibrosis (IPF) are important events in the disease trajectory of IPF, associated with punctuated decline in lung function with significant mortality and morbidity associated with it. These episodes are idiosyncratic, and often unpredictable and may have triggers. Our diagnostic criteria for these events, etiology, pathogenesis, risk factors and management continue to evolve over the years, with limited availability of qualitative research data to help guide management. Outcome in general is poor with no well-defined therapy but prevention may be possible with use of Nintedanib. Our chapter aims to explore the contemporary knowledge of the key aspects of this disease entity.

1. Introduction

Acute exacerbations of Idiopathic Pulmonary fibrosis (AE-IPF) represent important milestone in the disease course of IPF, which is the most common disease among the group of Idiopathic interstitial pneumonia (IIP). The IPF is more common among males and in the elderly age group [1]. Although the exact etiology of AE-IPF is unknown, there are many important risk factors as well as triggers that have been identified. There is associated accelerated decline in lung function which leads to poor prognosis [1]. It is estimated that about 35 to 46% of deaths in IPF are caused by AE-IPF [2]. In hospital mortality is more than 50% and follow up after hospitalization shows a mortality up to 73% at the end of 90 days [2, 3]. Although treatment with high dose Gluco corticoids have been used extensively, there is lack of controlled well designed trials to support its use and in fact survival has been shown to be decreased with steroids and or other immune suppressants [4, 5]. Some newer anti-fibrotic agents like Pirfenidone and Nintedanib may improve survival, the latter may be helpful in preventing AE-IPF [1, 6, 7].

2. Criteria for diagnosis of AE-IPF

Acute exacerbation of IPF is recognized with the help of a set of criteria laid out by the International IPF Working group network which is as follows [2, 4, 8, 9, 10, 11, 12].

“An acute, clinically significant respiratory deterioration characterized by evidence of new widespread alveolar abnormality”.

The following four diagnostic criteria have to be met as shown in Table 1.

A known or concurrent diagnosis of IPF

Clinical respiratory deterioration noted “typically” in the preceding 30 days.

Presence of typical UIP pattern on CT chest including bilateral basilar reticular changes with honeycombing and traction bronchiectasis. Superimposed ground glass attenuation and /or consolidation is necessary in exacerbation.; When possible specific UIP pattern may be combined with histopathological information to make a more robust diagnosis.

Absence of heart failure, Pulmonary embolism, fluid overload or any other differential pathology.

Note that endotracheal aspirate is not necessary as per new diagnostic criteria.

The 30-day time limit of clinical deterioration is not strictly enforced.

Exclude other causes of Interstitial Lung disease such as drug toxicity, connective tissue disease, hypersensitivity pneumonitis, etc.

Table 1.

Criteria for diagnosing IPF exacerbation as per working group idiopathic pulmonary fibrosis network (IPF net).

The guidelines also provided certain clarifications that help in making a diagnosis [11].

Events that are clinically considered to meet the definition of acute exacerbation of IPF but fail to meet all four diagnostic criteria owing to missing computed tomography data are to be termed “suspected acute exacerbations.” For example, if CT scan shows unilateral ground glass attenuation or data available is incomplete [2].

If the diagnosis of IPF is not previously established, this criterion can be met by the presence of radiologic and/or histopathologic changes consistent with usual interstitial pneumonia pattern on the current evaluation [11].

It is to be noted that the term “idiopathic” was removed from the older definition, as it was seen to be restrictive [11]. Making a distinction between idiopathic and non-idiopathic respiratory events is not easy as there are not well defined clinical or biological criteria [11]. So, although the acute deterioration could be due to an infectious etiology and it is not necessary to rule this out for the purpose of definition, at a practical level infection needs to be diagnosed and treated empirically or definitely as it does have a definite therapeutic recourse [11, 13]. It also follows from this that Broncho alveolar lavage (BAL) is not needed for diagnosis and hence it will help capture more of such events, but at the expense of specificity [11, 13]. BAL may not be needed when HRCT pattern is consistent with UIP, but when the Usual interstitial pneumonia (UIP) pattern is indeterminate or suspect then BAL can be useful [12]. Similarly, surgical lung biopsy (SLB) is recommended only when UIP is indeterminate or suspect [12]. However there is considerable morbidity and mortality associated with BAL or surgical biopsy in the context of AE and hence such procedures are to be generally avoided [11].

Similar to Acute lung injury in non-IPF lungs precipitated by triggers such as aspiration, post-operative, medication etc., exacerbation in IPF can be sub-categorized as either “triggered” when a known precipitating etiology is documented or “Idiopathic” when no such etiology is apparent [11].

The new definition also replaced the 30-day time restriction for the acute deterioration to “typically or generally of less than one-month duration” [11]. The phrase “typically less than 1 month” was included to provide precision but allow for the inclusion of exceptions that clinicians believe represent acute exacerbations [11]. A more flexible time interval may lead to “heterogeneity” among clinicians and clinical trial endpoint definitions for acute exacerbation [11].

3. Epidemiology

The incidence of acute exacerbations is variably reported in literature. This is because exacerbations could be more common in certain populations like more elderly people who are also likely to have severe disease, inconsistent definitions and its use, statistical design, follow up time and other factors [2, 9]. Exacerbations are less common in mild to moderate disease compared to severe disease [3, 14]. Reporting can vary depending on the type of study as well. Prospective trials may lack sufficient data to report all exacerbations [2, 4, 8], typically include younger patients with less comorbidities, with mild to moderate disease and therefore may under report incidence [15]. Retrospective studies may over report depending on the criteria used, by including events with pulmonary embolism, heart failure etc. [2, 13, 16].

Suspected exacerbations, which may not satisfy the definition of definitive AE-IPF are also important as they are associated with poor outcomes [4].

A meta-analysis analyzed six trials and reported acute exacerbation rate of 41 acute exacerbations per 1000 patient/years [14]. Rate of acute exacerbations were much lower in trials that included only mild to moderate disease [14].

In a retrospective study from Korea, which included 461 patients with IPF of which 269 cases were biopsy-proven and the median follows up period was 22.9 months, acute exacerbation occurred in 96 (20.8%) patients, and 17 (17.7%) of those acute exacerbation patients experienced multiple episodes of acute exacerbations (range 2–3 episodes). The incidence of acute exacerbation was noted to be 14.2, 18.8 and 20.7 percent at the end of 1, 2 and 3 years respectively [17].

It is to be noted that exacerbation of IPF may even occur in individuals with limited fibrosis and well-preserved lung function [2]. In the STEP-IPF trail, definite AE-IPF was reported as 40 per 1000 patient-years. However, the combined definite and suspected AE-IPF increased the exacerbation rate to 200 per 1000 patient-year [2, 4, 8, 13].

There have been reports of increased rate of AE-IPF in people of Asian descent in the far east such as Japan and Korea, however this has not been proven by randomized control trials [7, 8, 18].

4. Risk factors and pathophysiology

There are many risk factors for acute exacerbation. However, the most important risk factor is advanced disease [2, 14]. Other factors described in literature include low Forced vital capacity [8, 17], recent decline in FVC [19, 20], Low diffusion capacity for carbon monoxide [4], low 6 min walk test [4], pulmonary hypertension [21], poor baseline oxygenation [4, 22], increased dyspnea score [4], younger age group [2], presence of concurrent coronary artery disease [4], higher body mass index [22], previous history of acute exacerbation of IPF [19, 23].

Some important triggers for acute exacerbation have been described.

4.1 Infection

Infections are very common causes of respiratory deterioration. Exacerbations are more common in winter and spring season [24] and in those who are immunosuppressed [2, 25]. Postmortem analysis, multiplex polymerase chain reaction (PCR), pan viral micro array, high output cDNA sequencing and other techniques have demonstrated that infectious etiology is incriminated in many but not all acute exacerbations [23]. Based on the cumulative evidence which demonstrates infection to be causing some but not all acute exacerbations, it is thought to be an important but not exclusive trigger for precipitating acute exacerbations.

4.2 Silent aspiration of gastric contents

Aspiration of gastric contents has been postulated to be a causative factor for IPF and exacerbations in IPF.

In a case control study involving 24 acute exacerbations and 30 controls, Pepsin level in Broncho alveolar lavage (BAL) was found to be fairly commonly present, suggestive of gastric aspiration being fairly common in IPF.8 of the 24 acute exacerbation of IPF patients had very high levels of Pepsin suggesting that aspiration of gastric contents could be a contributor for acute exacerbation [26].

In a study involving 32 patients with asymmetric idiopathic pulmonary fibrosis (AIPF) compared with 64 matched controls with symmetrical IPF, Gastro esophageal reflux disease (GERD) and AE-IPF was significantly higher in patients with AIPF with the left side being less commonly involved [27].

On the contrary, in a post hoc analysis of the two Phase III randomized placebo-controlled INPULSIS trials of Nintedanib in patients with IPF, the rate of decline of FVC in the placebo group was much higher in patients who were taking an antacid (Proton pump inhibitor or H2-receptor antagonists) at baseline when compared to those who were not [difference of − 47.5 mL/year (95% CI: −105.1, 10.1); p = 0.1057] [28].

The data as it is apparent, that although gastro esophageal reflex has been widely debated to be causative, is not very definitive. Therefore, it can be likely that the aspiration of gastro esophageal contents can trigger acute exacerbations like infections but is not the sole causative factor.

4.3 Surgery and other interventions

Many surgical procedures like bronchoscopy and BAL, lung biopsy, lung resection, non-thoracic surgery and others can precipitate acute exacerbation [24, 29, 30, 31, 32]. The mechanism of action is unclear but could be related to stress like volutrauma, barotrauma, free oxygen radicals, or intra operative fluid balance.

4.4 Air pollution

Air pollution can be a cause for interstitial lung disease. In a retrospective south Korean longitudinal cohort, out of 505 IPF patients 436 patients were included in the final analysis. 75 patients experienced at least one exacerbation. There were 89 acute exacerbation events occurring over 1699 patient-years, for an incidence rate of 5.2 exacerbations per 100 patient-years [23].

Air pollution data for each of the five pollutants Ozone (O3), Nitrogen di oxide (NO2), particles with a 50% cut-off aerodynamic diameter of <10 μm (PM10), sulfur dioxide (SO2) and carbon monoxide (CO) were measured prospectively at Tele-Monitoring-Systems (TMS) situated throughout Korea. Each TMS recorded hourly measurements of each pollutant during the study period. Mean and maximum exposures of all these 5 pollutants were recorded over the 42-day period prior to the exacerbation period. Acute exacerbation of IPF was significantly associated with important measurement metrics of O3 and NO2 during the exposure period. Mean Ozone and Nitrogen dioxide levels were weakly correlated; however, both were statistically significant independent predictors of AE-IPF [23].

4.5 Medications

Medications can provoke respiratory deterioration in interstitial pneumonia which closely resembles acute exacerbation. Such drugs include everolimus, interferon-gamma and others [33, 34]. Drugs and surgery used to treat Lung cancer patients with interstitial pneumonia did not appear to cause more exacerbations compared to best supportive care and hence should not be withheld when treating Lung cancer with interstitial pneumonia patients [35].

As noted, there have been many triggers associated with acute exacerbation of IPF. However currently the most accepted theory is that exacerbation is thought to be “an acceleration of the underlying inflammatory fibro proliferative disease process”. This theory is supported by markers of cell injury as well as genetic expressions.

In one study by Collard et al., 47 patients with acute exacerbation of IPF, 20 patients with stable IPF and 20 patients with acute lung injury were studied. Plasma from these patients were collected and measured for biomarkers of cell activity/injury-receptor for advanced glycation end (RAGE) products, surfactant protein D, KL-6, von Willebrand factor; systemic inflammation-Interleukin-6; and biomarkers of coagulation/fibrinolysis-protein C, thrombomodulin, plasminogen activator inhibitor-1. Plasma from patients with AE-IPF showed higher levels of markers of type II alveolar epithelial cell injury/proliferation, endothelial cell injury, and coagulation/fibrinolysis very much like stable IPF but the response was much more exaggerated. This biomarker profile was different from patients with acute lung injury which was consistent with type I alveolar epithelial injury [36].

In another study, RNA was extracted from 23 stable IPF lungs, 8 IPF lungs with acute exacerbation of IPF and 15 control lungs. The gene expressions were studied. Results indicated that 579 genes were differentially expressed between stable IPF and acute exacerbation of IPF. Functional analysis of these genes was not suggestive of infectious or inflammatory etiology. Gene expression patterns in acute exacerbations of IPF and IPF samples were quite similar and different from the control lung arm [37].

Other immunological theories have also been proposed. Annexin 1 is an antigen found in human body which is increased in patients who have AE of IPF [38]. This antigen can induce both humoral and cell mediated immune responses and certain parts of this antigen have been implicated in the pathogenesis of AE of IPF [38]. Certain molecular studies have also been performed. Heat shock protein 47 (HSP47), has been studied and found to be a good bio marker for collagen production and secretion [39]. In studies comparing stable and AE-IPF patients, serum HSP47 were significantly elevated in AE-IPF patients in comparison to stable IPF patients [39]. Ironically patients who have anti-HSP70 autoantibodies in smaller studies have much poor prognosis due to AE of IPF when compared to controls or even those patients who have IPF but negative anti-HSP70 autoantibodies [40, 41].

Epithelial damage and impaired healing by abundance fibrosis has been an important theory that tries to explain the pathologic damage in IPF patients. When compared to IPF patients, patients with AE of IPF have higher bio markers of neutrophilic damage such as Alpha-defensins which are produced by activated neutrophils [37, 42] and also increased levels of Fibrocytes have been noted which have been found to be associated with worser outcomes in patients with AE of IPF [42, 43].

5. Clinical features

Patients present with worsening respiratory symptoms generally which are less than 30 days in duration. It consists of cough, worsening dyspnea especially on exertion, fever, malaise, and other flu like symptoms. Criteria for diagnosis include PaO2/FiO2 ratio < 225 or a decrease in PaO2 of ≥10 mmHg over time [8, 16].

Physical examination is consistent with IPF including bibasilar crackles on auscultation, but with increased respiratory rate [37].

Laboratory testing and imaging are directed to rule out other differentials like congestive heart failure, Myocardial infarction, pulmonary embolism, pulmonary hypertension, pulmonary infections etc. [2, 8]. Accordingly, complete blood count, B-type natriuretic peptide, C-reactive protein, chemistry profile including Blood urea nitrogen and serum creatinine can be performed along with highly sensitive troponins. Laboratory values are consistent with an infectious or inflammatory process but there is no evidence of infection on Blood culture, Urine antigen tests or Broncho alveolar lavage (BAL) if these tests are undertaken [8]. BAL if performed typically shows neutrophilic predominance [8]. BNP is typically elevated in heart failure and pulmonary hypertension [44]. Echo may be beneficial in heart failure and pulmonary hypertension [44]. CRP is typically elevated in infections and inflammation [44]. Pro calcitonin can guide when infection is suspected and even helping with limiting duration of antibiotic use [45]. D-dimer and CT pulmonary angiogram can help with ruling in or ruling out Pulmonary embolism [44].

High resolution computed tomography (HRCT) reveals bilateral ground glass or consolidative opacities superimposed on a background of typical HRCT features of IPF which includes bibasilar reticular opacities, honeycomb changes, and traction bronchiectasis [2].

Acute exacerbation of IPF is essentially a clinical diagnosis aided by predominantly noninvasive test. Although surgical biopsy can be performed for diagnosis which may show diffuse alveolar damage, the mortality and morbidity in the acute situation appear to be prohibitively high and not recommended [46].

6. Treatment and prognosis

Treatment is primarily supportive in nature.

Supplemental oxygen consisting of low flow and high flow oxygen can be used to keep Oxygen saturation (Spo2) more than 92%. Noninvasive ventilation mechanical ventilation (NIMV) and Mechanical ventilation (MV) is used as needed.

In a retrospective review of 19 hospitalized patients with IPF and AE, 1/3rd of patients had an infectious etiology, the percentage of patients who were discharged alive was 37% and only 14.8% of patients were alive at 1 year [47]. Patients with IPF experience AE very commonly. In IPF, about 40% of patients may die due to AE [48, 49]. In another observational study of 112 patients, 56 patients (42.9%) died due to AE [48]. The five-year survival rate of all patients with IPF was 38.3% and the Median survival time was 3.1 years post diagnosis. However, in patients who had an AE, the five-year survival rate was 10.7% and median survival time was 0.6 years [48].

In a pooled data consisting of nine studies, including 135 patients who were intubated for AE of IPF the cumulative mortality was 118 (87%) and short-term mortality (within 3 months of discharge) was 127 (94%) [50]. Therefore AE of IPF is not only common, but also a very poor predictor of survival. Based on these observations, the 2011 IPF guidelines discouraged MV in the vast majority and recommended its use in a selective minority of patients after careful weighing of risks and benefits [11]. However, this data pertains to a time period before 2007. In a study that reported US national data of 1703 patients who received Mechanical ventilation (MV)and 778 patients who received Noninvasive mechanical ventilation (NIMV), mortality was about 50% in those who received MV compared to 30% for those who received NIMV. The mortality of IPF patients treated with MV improved from 58.4% in 2006 to 49.3% in 2012 which was significant [51]. Overall this is suggestive that survival of patients treated with MV has seen a marginal improvement which could be due to various factors such as relative changes in diagnostic criteria and their use, difference in variables used and study design, differences in the severity of disease (patients with decreased FVC have poorer prognosis), judicious selection of patients who were placed on MV and widespread adoption and use of lung protective ventilation strategies [52]. Hence it is imperative that well informed discussions relating to advance care directives are made at the time of diagnosis and re visited when hospitalized [50]. In patients who are candidates for Lung transplant, the use of mechanical ventilation and extra corporeal membrane oxygenation, can be effective and lifesaving [11, 52, 53, 54, 55].

Ventilator induced lung injury (VILI) secondary to use of MV results in lung damage and poor prognosis [52]. In IPF, the lungs are fibrotic and non-compliant. Lower PEEP may be more beneficial and protective along with low tidal volume ventilation, hence minimizing both volutrauma and barotrauma [52, 56]. NIMV and high flow oxygen are being increasingly used and may be beneficial [56, 57]. Both NIMV and High flow Oxygen could be beneficial in patients who are not appropriate or choose to forego intubation, the survival may be the same with both modalities [57, 58], with high flow nasal oxygen being better tolerated, allowing patients to even eat and drink [57, 58, 59]. Hence they could be very effective means for palliative care [57, 58].

In patients who undergo thoracic surgery, VILI may be a potential etiological mechanism and can be minimized by the aforementioned lung protective ventilation including reducing lung volume, low PEEP, low partial pressure of inspired oxygen (Fio2), and less invasive surgical techniques [52, 56, 60, 61].

Symptoms such as dyspnea are treated with a palliative intent [8, 11]. Oxygen and opioids can also be given for symptom control [8, 11].

In IPF, a combination of inflammation, epithelial cell injury, fibro proliferative repair, and tissue remodeling which interact with the coagulation system help characterize IPF as a procoagulant state [62]. The use of therapeutic anticoagulation such as Warfarin or Alfa-Thrombomodulin has proven to be either harmful or non-beneficial in well conducted studies [63, 64, 65]. There is no evidence supporting therapeutic anticoagulation in patients experiencing acute exacerbation [63]. Nevertheless, patients with IPF have almost twice the risk of venous thromboembolism compared to general population and hence pharmacological venous thromboprophylaxis should be routinely used in hospitalized patients [66, 67].

There is not enough evidence of protective role from the use of antacids, but patients who are already using them can continue with their use [68, 69]. Evidence is often contradictory if antacids protect or may potentiate or worsen AE of IPF [70, 71].

Corticosteroids have been used extensively but the practice is driven by expert opinion and anecdotal reports and not driven by good data. Expert guidelines give a weak recommendation to the use of steroids [2, 6]. Acute IPF is characterized by high degree of inflammation with areas of diffuse alveolar damage secondary to Acute Lung injury and organizing pneumonia [9]. Therefore, the use of high dose steroids is intuitively thought to be beneficial [9, 72, 73, 74], in spite of absence of good data from randomized control trials [9]. Dosage and duration are also not well defined in literature, although it is typical to use initially high dose corticosteroids followed by a rapid tapering course, as longer duration of steroids may be harmful in IPF [9, 62]. In EXAFIP, a randomized control trial comparing Cyclophosphamide with corticosteroids against placebo with corticosteroids in AE of IPF, the steroid regimen used in all patients was Methylprednisolone 10 mg/kg per day for 3 days followed by a progressive taper to 10 mg per day for patients above 65 kg and 7·5 mg per day for patients below 65 kg at the end of 6 months [75]. Similarly in a RCT involving 77 patients in Japan, Alpha-Thrombomodulin (ART-123) was compared against placebo. All patients received glucocorticoids in two courses of pulse Methylprednisolone (500–1000 mg/day) for 3 days followed by Prednisolone 0.5–1.0 mg/kg/day for 4 days followed by gradual taper [64]. Smaller retrospective studies have shown that using high dose of glucocorticoids used in first 30 days prevent recurrence of exacerbation when compared to lower doses in the same duration or after 30 days but this has not been substantiated by other studies [76, 77]. The use of high dose steroids has been noted to increase survival in non-IPF exacerbation of Interstitial lung disease [76]. It is noteworthy that some studies and even guidelines recommend using no immunosuppressives in select patients, as mortality was no better in the immunosuppressed group when compared to the non-immunosuppressed, with higher incidence of infection in the immune suppressed group, especially in severe disease [11, 14].

Concomitant Immunosuppressive therapy with steroids have also been used and the evidence base for this practice is also not very sound [2, 48]. While treatments such as Alfa-Thrombomodulin and Cyclophosphamide along with concomitant glucocorticoids have been subjected to randomized control trials and have not been shown to improve outcomes [64, 75], others do not have much evidence as they were too small, were uncontrolled, used historical data as control or had no control arm [2]. The latter studies have used medications like Tacrolimus, Cyclosporin-A, Rituximab combined with plasma exchange, Intravenous Immunoglobulin, and polymyxin B-immobilized fiber column (PMX) [6]. Other medications that have been used include Acetylcysteine as standalone therapy, sildenafil, bosentan, interferon-gamma 1b, warfarin, ambrisentan, and imatinib [8].

In a small retrospective study consisting of 11 patients in each group, Corticosteroids alone were compared with Cyclosporin-A and Corticosteroids. The mortality was similar in each group but the Cyclosporin-A group appeared to have longer survival [78]. However in a larger retrospective review with 384 patients in Cyclosporin-A and high dose Corticosteroids and 7605 patients treated with high dose Corticosteroids alone in Japan, no change in survival was noted [79].

Other considerations include empiric treatment of a course of antibiotics since infectious etiology can be treated but cannot be ruled out conclusively in the vast majority of cases [2]. Procalcitonin has been used in clinical trials and can reduce the duration of antibiotics (8.7 ± 6.6 compared to 14.2 ± 5.2 days in the routine treatment group), without any effect on treatment success, mortality rate, days of hospitalization and ventilation therapy [45]. Tacrolimus, an immunosuppressive drug used widely in solid organ transplant patients including Lung transplant was found to have beneficial survival effects and protection against future exacerbations in small retrospective studies, with lack of data from better designed controlled studies [80]. Direct hemoperfusion with Polymyxin B immobilized fiber column (PMX-DHP) has been used in AE of IPF to absorb endotoxins and reactive oxygen species, among other toxic substances, as well as selectively remove activated neutrophils and preventing activation of monocytes with a goal to limit endothelial damage [81, 82]. PMX-DHP may act by adsorbing harmful cytokines such as vascular endothelial growth factor and may have anti-fibrotic effect [83, 84]. The adsorption of proinflammatory, profibrotic and proangiogenic cytokines is postulated to be an important mechanistic action of PMX-DHP [83]. The use of PMX-DHP along with Corticosteroids has demonstrated improvement in oxygenation, with possible improvement in survival in a multicentric Japanese retrospective study with 73 patient who had AE of IPF [82]. There is a prevailing hypothesis that auto antibodies may have a role in IPF progression. Removal of these auto antibodies by plasma exchange and Rituximab followed by IVIG subsequently may be beneficial in AE-IPF [62, 85]. A small pilot study involving 11 patients has shown the safety and possible efficacy, paving way for a Phase 3 randomized control trial [85].

Interestingly in a retrospective study, patients who were not on any immunosuppression had better survival than those who were on immunosuppression [5].

7. Prognostic score

There has been considerable interest in developing prognostication scores for AE of IPF. A number of markers have been used in different studies and Forced vital capacity %, Diffusion capacity for Carbon monoxide, Pao2/Fio2 (P/F) ratio, HRCT patterns, Acute Physiology and Chronic Health Evaluation II score (APACHE II), Glasgow prognostic score and serum biomarkers like C-reactive protein (CRP), Krebs von den Lungen-6 (KL-6) have all been considered [86]. In a retrospective study of 108 patients, a lower FVC % at baseline (1 year before AE) and P/F ratio on AE presentation were predictive of mortality [86]. In another study of 103 AE-IPF cases, a combination of P/F ratio less than 250 (P), CRP ≥ 5.5 (C), and diffuse HRCT pattern (radiological) (R), together called as PCR index was used to stratify and predict mortality at the end of 3 months [87]. In a systematic review and meta-analysis, 37 studies and 31 prognostic factors were analyzed [88]. Five independent variables after multivariate analysis were found to be helpful with prognostication namely APACHE II score, P/F ratio, LDH level, white blood cell (WBC) count, and oxygen therapy before AE [88]. Interestingly the latter did not find use of FVC or imaging scores to be helpful in terms of prognostication [88]. Prognostication scores and models are certainly good research tools but not commonly used in clinical practise as no intervention other than good supportive care has been found to be useful.

8. Prevention

Prevention of exacerbation in IPF is the most effective strategy as we do not seem to have very effective therapies once the exacerbation gets underway. Avoidance of air pollutants [23], preventing infections like Streptococcal pneumonia and Influenza by vaccination [26, 61], general hygiene measures like handwashing again to prevent infections [52], and judicious use of antacids may be helpful strategies [68, 69].

Many medications have been tested, using prevention of acute exacerbation as an end point. Acetylcysteine monotherapy, bosentan, interferon-gamma, sildenafil, showed no effect [8]. Others like imatinib, ambrisentan, triple therapy (prednisone, azathioprine, acetylcysteine combination) and warfarin showed increased risk of exacerbation [8].

Azuma et al. studied 107 IPF patients in a phase 2 Randomized placebo-controlled trial comparing Pirfenidone and placebo. Although there were no acute exacerbations noted in the Pirfenidone arm compared to placebo [89], the same results could not be reproduced in a phase 3 RCT with 275 patients, showing no difference between the intervention and control arm [18]. In the large phase 3 RCT’s, CAPACITY and ASCEND which compared Pirfenidone with placebo yet again, unfortunately AE of IPF as an end point was not studied [90, 91]. Nevertheless, a pooled analysis of the CAPACITY and ASCEND trial did reveal a reduction in non-elective respiratory related hospitalization favoring Pirfenidone [92]. Interestingly Pirfenidone in small studies has proven to be safe and effective in preventing exacerbations in peri operative period in patients who were given 2–4 weeks of medication prior to surgery and continued post operatively when compared against historical controls [93, 94]. Larger RCTs need to be performed for this promising intervention [94].

Another antifibrotic agent Nintedanib was studied after Pirfenidone, which showed a favorable effect against placebo for preventing AE of IPF in the phase 2 TOMORROW trial and phase 3 INPULSIS-2 trial, but no such effect was seen in the phase 3 INPULSIS-1 trial [95, 96]. However the pooled analysis of patients from TOMORROW and INPULSIS trials [6], consisting of 1231 patients (Nintedanib n = 723, placebo n = 508), the hazard ratio for time to first acute exacerbation was 0.53 (95% CI: 0.34, 0.83; p = 0.0047) favoring Nintedanib. The proportion of patients with ≥1 acute exacerbation was 4.6% in the Nintedanib group and 8.7% in the placebo group [6]. Nintedanib can be added after recovering from an exacerbation or continued if it was previously being used.

In a systematic review and meta-analysis, 12,956 patients were included comparing the use of anti fibrotics (Pirfenidone or Nintedanib) vs. nonuse of antifibrotics, which showed that the use of antifibrotics decreased all-cause mortality, RR 0.55 (95% CI, 0.45–0.66). The same review included seven studies involving 2002 treated and 1323 non-treated patients, and showed a decrease in AE, which was statistically significant for Nintedanib (RR 0.62 [95% CI, 0.43–0.89) but only non-significant decrease for Pirfenidone, RR of 0.57 (95% CI, 0.29–1.12) [1].

Overall, the evidence favors Nintedanib over Pirfenidone in terms of preventing AE of IPF. However, there are no head to head comparisons between these two approved medications and real world data could produce results to the contrary [97]. Hence it would be prudent to plan design and conduct appropriate RCT that would give an unambiguous answer to this very important question.

9. Conclusion

Episodes of Acute exacerbation are important events in the disease course of IPF. Up to 40% of deaths in IPF are caused by acute exacerbations. After the initial diagnosis, the median survival of patients with acute exacerbation was much shorter (15.5 months) than that of patients without respiratory deterioration (60.6 months). The 5 year rate of survival of patients with acute exacerbation was 18.4%, whereas 50.0% of patients without respiratory deterioration survived.

While medications like Nintedanib can slow down progression of disease and prevent exacerbations, once diagnosed it has no known effective treatment. Hence more research is needed to alter the disease course of IPF as well as prevent the occurrence of these exacerbations which invariably is an indicator of poor prognosis.