High-dose steroid therapy (steroid pulse therapy) was developed originally for the control of host rejection to organ transplants and the treatment of collagen disease [6]. Subsequently, it began to be used also for the treatment of AE-IPF. Global guidelines (official ATS/ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis) justify high-dose steroid therapy for AE-IPF, although the evidence level is low [7]. Japanese guidelines recommend 3-day treatment with 1 g methylprednisolone daily.

According to the paper published by Kondoh et al., the first report on this therapy, treatment of 3 AE-IPF patients with methylprednisolone 1 g for 3 days followed by tapering of the steroid dose led to alleviation of the clinical symptoms and improvement of the PaO₂/FiO₂ (P/F) ratio [2]. However, Song et al. reported that following administration of high-dose steroid therapy to 13 of 90 patients with AR-IPF, only 7 survived, and there was no remarkable improvement of the survival rate [8]. In the patient series reported by Al-Hameed et al., of the 25 patients with AE-IPF to whom high-dose steroid therapy was applied (in combination with cyclophosphamide in 8 patients), 24 died [9]. Usui et al. conducted a retrospective survey of 52 patients with acute exacerbation of interstitial pneumonia (including cases of IPF) and concluded that high-dose steroid therapy did not improve the survival rate [10]. On the other hand, according to the report by Takahashi et al., who analyzed 17 cases of AE-IPF receiving high-dose steroid therapy, the mean survival time (MST) was 1.2 months in patients showing no improvement of the alveolar-arterial oxygen difference (A-aDO₂), while the MST was prolonged to 4.5 months in patients showing transient improvement of the A-aDO₂ and to 24.4 months in patients showing sustained improvement of the A-aDO₂ for 3 months, thus concluding that improvement of the A-aDO₂ may be a useful predictor of the responses to high-dose steroid therapy [11]. Thus, all previous reports on the outcome of high-dose steroid therapy for AE-IPF are based on retrospective surveys, and no RCT has been carried out. Furthermore, high-dose steroid therapy is associated with a high risk for adverse reactions and complications, and there is no consensus on whether it should be applied definitely [12–14].

AE-IPF is characterized by organizing DAD as a pathologic feature resembling that in ARDS. Therefore, an RCT of steroid therapy for ARDS may provide information useful for the treatment of AE-IPF [15–17]. However, even though a similarity of the pathologic changes is noted, ARDS without preexisting lung lesions and ARDS attributable to extrapulmonary factors should be considered as conditions differing from AE-IPF in terms of the mechanism of onset.

In the RCT reported by Bernard et al., treatment with methylprednisolone (30 mg/kg) every 6 h during the acute period after the onset of ARDS resulted in no difference in the death rate from that in the placebo group and no improvement of the clinical indicators [18]. In the RCT reported by Annane et al., hydrocortisone 50 mg was administered every 6 h and 9-α-fludrocortisone 50 μg was administered once daily, both for 7 days soon after the onset of ARDS [19]. The length of time until weaning from respiratory support was longer in the steroid therapy group than that in the placebo group; however, there was no difference in the death rate between the two groups. In the RCT conducted by ARDSnet, treatment with medium-dose methylprednisolone (dose level reduced gradually from 2 mg/kg/day) from Day 7 to Day 28 after the onset of ARDS resulted in no difference in the death rate from that in the placebo group. Weaning from respiratory support was achieved earlier in the methylprednisolone treatment group, although auxiliary ventilation was often needed after weaning. The death rates at 60 days and 180 days were higher in the group that had begun to receive methylprednisolone on Day 14 after onset than that in the placebo group [20].

In contrast to these reports, Meduri et al. reported that treatment with methylprednisolone soon after the onset of ARDS at a medium initial dose level (1 mg/kg/day, gradually reduced over time) resulted in shortening of the duration of mechanical ventilation and length of stay in the ICU as compared to the placebo group, associated with a reduction of the death rate [21]. The same investigators reported that treatment with methylprednisolone from Day 7 onward after the onset of ARDS at gradually reducing dose levels from 2 mg/kg/day resulted in improvement of the lung injury score (LIS), P/F ratio, and multiple organ dysfunction score (MODS) and reduction of the death rate as compared to the placebo group [22]. A systematic review summarizing these reports concluded that treatment of ARDS with methylprednisolone at low to medium dose levels (0.5–2.5 mg/kg) initiated within 14 days after the onset can lower the death rate without increasing the incidence of adverse reactions [15, 16].

At present, high-dose steroid therapy (methylprednisolone 1 g/day, 3 days) is often used for the treatment of AE-IPF. However, if data from RCTs on ARDS are taken into account, it would seem rational to administer methylprednisolone at low to medium dose levels (0.5–2.5 mg/kg) in the acute stage or within 14 days after the onset of ARDS as another alternative of treatment.

Immunosuppressants such as cyclosporine A, tacrolimus, cyclophosphamide, and azathioprine are often used in combination with steroids for the treatment of AE-IPF. Cyclosporine A is a cyclic polypeptide antibiotic produced by fungi and has been shown to exert immunosuppressive activity through inhibition of interleukin-2 (IL-2) production by helper T cells and other mechanisms. According to a retrospective study conducted by Inase et al., 4 of 7 patients with AE-IPF survived following combined therapy with high-dose steroid (methylprednisolone 1 g/day, 3 days) + cyclosporine A (1.0–2.0 mg/kg, trough 100–150 ng/ml), while all of the patients treated with high-dose steroid therapy alone without concomitant cyclosporine A treatment died [23]. In a retrospective study conducted by Homma et al., the MST was 1.7 months in the prednisolone-alone treatment group (n = 35), shorter than the 9.9 months recorded in the combined prednisolone + cyclosporine A (50–200 mg) treatment group (n = 9) [24]. In a retrospective study reported by Sakamoto et al., comparison of the prognosis between 11 AE-IPF patients treated with high-dose steroid therapy alone and 11 AE-IPF patients treated with high-dose steroid + cyclosporine A (1–2 mg/kg) therapy revealed a longer MST in the combined treatment group (502 days) than in the uncombined treatment group (60 days) [25]. However, a retrospective study by Okamoto et al. revealed no difference in the MST between the 8 AE-IPF patients treated with steroid + cyclosporine A therapy and 9 AE-IPF patients treated with a steroid alone [26].

Cyclophosphamide is an immunosuppressant agent that is used for the control of host rejection of organ transplant and treatment of collagen vascular diseases. According to Japanese guidelines, repetition of treatment with cyclophosphamide (500 mg/day) in combination with a steroid at intervals of 1–2 weeks is recommended in the treatment of AE-IPF, although the evidence level is low. As an example of cyclophosphamide therapy for AE-IPF, there is one report of a patient with AE-IPF developing after influenza vaccination, whose life was successfully saved by treatment with a combination of a high-dose steroid (methylprednisolone 1 g/day, 3 days), cyclophosphamide (500 mg/day), sivelestat, and polymyxin B-immobilized fiber column hemoperfusion [27]. However, according to the report by Ambrosini et al., 4 of the 5 AE-IPF patients who received treatment with high-dose steroids in combination with cyclophosphamide or azathioprine died 13 days after the start of treatment, on average [28]. Also according to the report by Parambil et al., all of the 7 AE-IPF patients (including 2 patients receiving treatment with steroids + cyclophosphamide) died [29].

Tacrolimus is a drug that suppresses the proliferation and differentiation of cytotoxic T lymphocytes through inhibition of IL-2 formation. Horita et al. reported a retrospective study in which combined administration of steroids with tacrolimus served as an effective strategy for treating AE-IPF [30]. In their study, all patients received high-dose steroid therapy (methylprednisolone 1 g/day, 3 days), with some additionally receiving serial intravenous infusion of tacrolimus for 5–14 days (target blood level 20 ng/ml), followed by oral tacrolimus treatment (target blood level 2 ng/ml). The P/F ratio and blood lactate dehydrogenase (LDH) level improved, and the MST was prolonged in the combined treatment group (n = 5) as compared to the uncombined treatment group (n = 10) (more than 92 days vs. 38 days, p < 0.05).

IPF is characterized by enhanced blood coagulability due to vascular endothelial disorders [31–34]. Kubo et al. reported the results of an RCT of anticoagulant therapy using warfarin in IPF patients receiving oral steroid therapy (n = 56) [35]. During the observation period, 60 % of the patients were hospitalized because of exacerbation of respiratory failure, which was often attributable to AE-IPF. During the hospital stay, the anticoagulant therapy group received low-molecular-weight heparin in place of warfarin. The death rate due to AE-IPF was lower in the anticoagulation treatment group than in the anticoagulant drug-untreated group (18 % vs. 71 %, P = 0.008), and elevation of the blood D-dimer level was associated with the death rate. However, Simon-Blancal et al. reported that the prognosis of patients with AE-IPF was not improved by combined steroid + low-molecular-weight heparin treatment [36]. Furthermore, Noth et al. reported, based on an RCT in patients with IPF, a higher death rate in the warfarin treatment group as compared to that in the placebo group [37].

Thrombomodulin is known to have high mobility group box-1 (HMGB1)-inhibitory activity in addition to anticoagulant activity [38, 39]. In recent years, a number of reports have been published from Japan on the results of treatment with AE-IPF with recombinant human soluble thrombomodulin (rhTM) [40–42]. Taniguchi et al. reported that when 40 patients with AE-IPF were treated with a high-dose steroid (methylprednisolone 1 g/day, 3 days) + cyclosporine, accompanied by 6-day rhTM treatment (0.06 mg/kg/day) in 20 of these patients, the death rate at 3 months was lower in the rhTM-treated group than that in the rhTM-untreated group (HR = 0.17, P = 0.015) [40]. Isshiki et al. treated 42 AE-IPF patients with a high-dose steroid (methylprednisolone 1 g/day, 3 days), accompanied by 6-day rhTM treatment (0.06 mg/kg/day) in 16 of these patients [41]. The blood HMGB1 protein level on Day 7 of treatment was lower, and the survival rate at 3 months was higher in the rhTM-treated group than that in the rhTM-untreated group (69 % vs. 38 %, P = 0.03). Tsushima et al. reported that patients with AE-IPF had abnormalities of the clotting system, such as increased plasma levels of fibrinogen degradation products (FDP), thrombin-antithrombin complex (TAT), plasma-alpha2 plasmin inhibitor complex (PIC), and D-dimers, and that the death rate at 28 days was lower in the group treated with a steroid + rhTM (n = 20) than that in the group treated with a steroid alone (n = 6) (35 % vs. 45 %, p < 0.05) [42].

Sivelestat is a neutrophil elastase inhibitor that was developed in Japan. Although a meta-analysis failed to demonstrate the effect of sivelestat in lowering the death rate due to ARDS [43], an animal study revealed its effect in suppressing the progression of bleomycin-induced pulmonary fibrosis [44]. Because patients with AE-IPF have been shown to have high blood neutrophil elastase levels, sivelestat is expected to be useful in the treatment of AE-IPF. In a phase 3 study of AE-IPF carried out in Japan, treatment with sivelestat was shown to alleviate or improve the shortness of breath and P/F ratio [45].

Pirfenidone is an antifibrotic drug that was developed in Japan. In an RCT carried out in 107 patients with IPF, the incidence of acute exacerbation during the 9-month observation period was 14 % in the pirfenidone-untreated group but 0 % in the pirfenidone treatment group (P = 0.003) [46]. Pirfenidone has also been reported to additionally suppress acute exacerbations of IPF after lung cancer surgery [47].

Nintedanib is an intracellular signal inhibitor targeting multiple tyrosine kinases. Of the phase III trials conducted in patients with IPF, INPULSIS-1 did not reveal any difference in the time until onset of acute exacerbation after the start of treatment between the nintedanib treatment group and the placebo group (AE-IPF hazard ratio, 1.15; 95 % CI, 0.54–2.42; p = 0.67). However, INPULSIS-2 revealed a longer period of time until onset of acute exacerbation in the nintedanib group (AE-IPF hazard ratio, 0.38; 95 % CI, 0.19–0.77; p = 0.005). Analysis of the combined data from both trials revealed that the incidence of AE-IPF was 36 % lower in the nintedanib group than that in the placebo group, although this difference was not statistically significant (HR, 0.64; 95 % CI, 0.39–1.05; p = 0.08) [48]. However, some investigators pointed out problems with the statistical analysis method in these trials, stating that the effect of nintedanib in suppressing AE-IPF could be deemed as statistically significant [49].

The antibiotic macrolide has been reported to suppress acute lung injury [50]. In a study in which AE-IPF patients receiving high-dose steroid + azithromycin treatment (n = 20) were compared with those receiving high-dose steroid + fluoroquinolone treatment (n = 56), the number of patients who died was 39 (70 %) in the fluoroquinolone-treated group but only 4 (20 %) in the azithromycin-treated group (p < 0.001) [51]. In regard to the reasons for the suppression of death among the AE-IPF patients treated with azithromycin, the immunosuppressive and/or anti-inflammatory activity of azithromycin has been suggested.

PMX-DHP therapy was developed as a means of treating sepsis through adsorption of blood endotoxins [52]. It has also been shown to be useful in the treatment of ARDS caused by sepsis, with the mechanism of action involving adsorption and removal of not only endotoxins but also of some other harmful substances [53–55]. For example, it has been reported that there are no differences in the blood levels of tumor necrosis factor-alpha (TNF-alpha), IL-6, IL-8, and IL-10 before and after PMX-DHP therapy [54] and that PMX-DHP therapy lowered the blood levels of matrix metalloproteinase (MMP)-9, tissue inhibitor of metalloproteinase (TIMP)-1 and HMGB-1, and the urinary levels of 8-hydroxy-2′-deoxyguanosine (8-OHDG) level [55, 56].

Recently, reports on PMX-DHP therapy for AE-IPF have been published primarily from Japanese facilities. Noma et al. reported that of 2 patients with AE-IPF administered high-dose steroid + PMX-DHP therapy, both showed reduction of the blood levels of HMGB1, monocyte chemoattractant protein-1 (MCP-1), IL-8, and IL-6, but only one of the two patients could be saved [57]. Tachibana et al. reported that of the 19 AE-IPF patients administered high-dose steroid + PMX-DHP therapy, 9 survived, allowing reduction of the blood IL-7 level to be identified as a prognostic factor [58]. Oishi et al. applied high-dose steroid + PMX-DHP therapy to 9 patients with AE-IPF, reporting that the therapy resulted in a reduction of the blood levels of IL-9, IL-12, IL-17, platelet-derived growth factor (PDGF), and vascular endothelial growth factor (VEGF) and that the extent of improvement of the P/F ratio was correlated with the quantity of VEGF adsorbed on the column [59]. Abe et al. observed the column used for PMX-DHP therapy in AE-IPF patients and found adsorption of numerous activated neutrophils [60]. The same investigators additionally reported that PMX-DHP therapy of AE-IPF patients resulted in HMGB-1 adsorption on the column and reduction of the blood HMGB-1 level [61]. Enomoto et al. reported no changes in the blood levels of IL-6, IL-8, IL-10, neutrophil elastase, or HMGB-1 after PMX-DHP therapy as compared to the levels recorded prior to the therapy; however, the peripheral white blood cell count decreased markedly after the therapy [62].