Article ID: CR-23-0059
Background: For patients with advanced heart failure, palliative care, including opioids, is needed as a treatment for refractory dyspnea. However, little evidence has been reported on the efficacy and safety of opioids, and their use is not well established.
Methods and Results: We have introduced a protocol for the use of opioids for dyspnea in patients with advanced heart failure admitted to Saitama Citizens Medical Center. Following this protocol, differences in clinical variables and outcome were investigated between patients in whom opioids were initiated intravenously or subcutaneously (i.v./s.c. group; n=13) and patients in whom they were initiated orally (oral group; n=18). In a comparison of baseline characteristics, significantly more patients in the oral group had a history of hospitalization for heart failure within the past year, and significantly more patients were treated with dobutamine and tolvaptan. After initiation of opioid treatment, both groups showed improvement in dyspnea; however, serial changes in vital signs were significantly greater in the i.v./s.c. group. The survival rate was significantly higher in the oral group (P<0.0001), with 33% of patients discharged alive.
Conclusions: The clinical use of oral opioids using a single-center protocol is reported, suggesting that oral opioids may be practical and effective for dyspnea in patients with advanced heart failure.
Heart failure (HF) is a serious condition associated with high morbidity and mortality rates, and is one of the major public health problems in aging society.1–3 The progression of chronic HF eventually leads to Stage D, defined by the American College of Cardiology/American Heart Association as:
… patients with truly refractory HF who might be eligible for specialized, advanced treatment strategies, such as mechanical circulatory support, procedures to facilitate fluid removal, continuous inotropic infusions, cardiac transplantation, other innovative or experimental surgical procedures, or for end-of-life care such as hospice.4
Almost 90% of patients with advanced HF have been reported to experience dyspnea with minimal exertion or at rest, significantly limiting their ability to perform daily activities.5 Therefore, alleviating refractory dyspnea in advanced HF patients is an important issue because it improves quality of life.
Opioids are known to be effective for relieving dyspnea in cancer patients.6 Current Japanese HF guidelines recommend low-dose morphine for refractory dyspnea in advanced HF patients;7 however, the evidence supporting this is insufficient. In addition, most advanced HF patients are complicated by renal dysfunction, so the dose of morphine, which is metabolized by the kidney, needs to be controlled. The difficulty of dose adjustment of oral morphine preparations is a particular problem in clinical practice.
In Saitama Citizens Medical Center, we have introduced a protocol for the use of opioids for refractory dyspnea in patients with advanced HF. In this study, we investigated the real-world clinical use of opioids for refractory dyspnea in patients with advanced HF, with the aim of determining the differences between oral and intravenous/subcutaneous (i.v./s.c.) opioid administration.
In November 2019, a protocol for the use of opioids for refractory dyspnea in patients with advanced HF was introduced at Saitama Citizens Medical Center and approved by the internal ethics committee (Figure 1). In particular, the Japanese national health insurance system does not cover the use of opioids for patients with HF, so this issue was also thoroughly discussed, and such use of opioids was approved by the internal ethics committee. From the introduction of this protocol until December 2022, 31 patients with advanced HF used opioids according to this protocol. These patients had uncontrolled symptoms of dyspnea and fatigue, even at rest, despite appropriate treatment for HF. In all cases, the use of opioids was thoroughly discussed with patients, family members, and medical staff, and agreed by them. If opioid use was requested, it was introduced according to the above-mentioned protocol. If oral intake was possible, opioids were basically initiated as oral medication.
In-hospital protocol for opioid use for dyspnea in patients with advanced heart failure.
Codeine was used as the first choice of oral opioid, whereas hydromorphone was used in patients with very severe symptoms. Codeine 20 mg was administered 3–4 times a day, and titrated up to a maximum daily dose of 160 mg taking into account symptoms and renal function. When using hydromorphone, it was initiated using 2–4 mg extended-release hydromorphone once a day, with 1 mg immediate-release hydromorphone used as rescue medication. Only in cases when oral intake was difficult or impossible was i.v. or s.c. morphine infusion chosen. The dose of morphine infusion was adjusted according to creatinine clearance. If symptoms were not well controlled, we considered the use of additional midazolam infusion.
Data CollectionWe retrospectively reviewed each patient’s medical chart to obtain information regarding clinical variables, medical history, comorbidities, medication, information on opioids, and clinical course. Patients were divided into 2 groups according to the route of opioid administration at initiation: those initiated i.v. or s.c. (i.v./s.c. group) and those initially orally (oral group). The degree of dyspnea was assessed using the Japanese version of the Support Team Assessment Schedule (STAS-J) scoring system8 with information obtained retrospectively from medical charts. The symptom list of the STAS-J version includes dyspnea, with the severity of dyspnea in patients being rated on a 5-point scale from none (=0) to “severe and continuous overwhelming breathlessness, unable to think of other matters” (=4).8 Vital signs data and the symptom scale were investigated at the start of opioid administration, as well as 24 and 72 h later.
The study was approved by the Institutional Review Board for Human Investigation of Saitama Citizens Medical Center. The study conformed to the principles outlined in the Declaration of Helsinki.
Statistical AnalysesAll values are expressed as the mean±SD, median and interquartile range when non-normally distributed, or percentages. Continuous variables were analyzed using an unpaired Student’s t-test for normally distributed variables, and the Mann-Whitney U test for non-normally distributed variables. Categorical data were analyzed using Fisher’s exact test. Serial changes between the 2 groups were compared by univariate repeated-measures analysis of variance (ANOVA). Cumulative event-free rates were estimated using the Kaplan-Meier method and compared using the log-rank test. For all analyses, 2-tailed P<0.05 was considered statistically significant. All statistical analyses were performed using EZR (Jichi Medical University Saitama Medical Center, Saitama, Japan), which is a graphical user interface for R (R Foundation for Statistical Computing, Vienna, Austria).9
Since the introduction of the protocol (i.e., between November 2019 and December 2022), opioids for dyspnea were administrated to 31 patients with advanced HF (17 men, 14 women; mean age 84.4±11.1 years; range 39–94 years). Thirteen patients initially received morphine i.v. or s.c. (i.v./s.c. group), whereas 18 patients initially received codeine or hydromorphone orally (oral group). In the i.v./s.c. group, 11 patients underwent i.v. administration and 2 patients underwent s.c. administration; 2 patients required continuous midazolam infusion for the last 2 days. In the oral group, 9 patients initially received codeine orally, whereas 9 patients initially received hydromorphone orally. Overall, 4 of 9 patients who initially received codeine were switched to hydromorphone during the clinical course.
Table 1 shows the baseline characteristics in the 2 groups. The oral group included significantly more patients with a history of hospitalization for HF exacerbation within the past year (78% vs. 31%; P=0.01), tolvaptan use (44% vs. 8%; P=0.04), and the continuous infusion of dobutamine (50% vs. 8%; P=0.02). Serum B-type natriuretic peptide concentrations at admission tended to be higher in the oral than i.v./s.c. group, although this difference was not statistically significant (936.1 vs. 608 pg/mL; P=0.06).
| Oral group (n=18) |
i.v./s.c. group (n=13) |
P value | |
|---|---|---|---|
| Age (years) | 82.6±13.5 | 86.9±6.3 | 0.29 |
| Male sex | 9 (50) | 8 (62) | 0.72 |
| Body mass index (km/m2) | 19.4±3.4 | 21.9±4.3 | 0.08 |
| LVEF (%) | 45.1±21.4 | 48.7±16.7 | 0.62 |
| Etiology of HF | 0.32 | ||
| ICM | 6 (33) | 4 (31) | |
| Non-ICM | 5 (28) | 6 (46) | |
| Valvular heart disease | 4 (22) | 1 (8) | |
| Others | 3 (17) | 2 (15) | |
| Comorbidity | |||
| Hypertension | 5 (28) | 5 (38) | 0.7 |
| Diabetes | 5 (28) | 3 (23) | 1 |
| Stroke | 4 (22) | 2 (15) | 0.21 |
| CKD | 17 (94) | 12 (92) | 1 |
| Atrial fibrillation | 9 (50) | 6 (46) | 1 |
| Malignancy | 5 (28) | 4 (31) | 1 |
| HF hospitalization <1 year | 14 (78) | 4 (31) | 0.01 |
| Oral medication | |||
| ACEI/ARB | 6 (33) | 7 (54) | 0.29 |
| β-blocker | 5 (28) | 5 (38) | 0.7 |
| MRA | 14 (78) | 5 (38) | 0.06 |
| Loop diuretic (oral) | 17 (94) | 9 (69) | 0.13 |
| Furosemide (intravenous) | 16 (89) | 12 (92) | 1 |
| SGLT-2 inhibitor | 3 (17) | 1 (8) | 0.62 |
| Tolvaptan | 8 (44) | 1 (8) | 0.04 |
| Amiodarone | 2 (11) | 2 (15) | 1 |
| Dobutamine | 9 (50) | 1 (8) | 0.02 |
| Laboratory data | |||
| BNP (pg/mL) | 936.1 [587.6–2,268.4] | 608.0 [237.8–948.6] | 0.06 |
| Creatinine (mg/dL) | 2.01±1.01 | 1.85±1.33 | 0.71 |
| eGFR (mL/min/1.73 m2) | 28.2±15.8 | 35.5±20.9 | 0.28 |
| Albumin (g/dL) | 3.34±0.40 | 3.08±0.64 | 0.19 |
| Sodium (mEq/L) | 136.1±5.7 | 139.0±5.2 | 0.16 |
| Hemoglobin (g/dL) | 11.8±2.7 | 11.4±2.5 | 0.66 |
Unless indicated otherwise, values are expressed as mean±SD, median [interquartile range], or n (%). ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; BNP, B-type natriuretic peptide; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HF, heart failure; ICM, ischemic cardiomyopathy; LVEF, left ventricular ejection fraction; MRA, mineralocorticoid receptor antagonist; SGLT-2, sodium glucose cotransporter 2.
The details of opioid use and patient outcomes are presented in Table 2 and Figure 2. Although there were no significant differences in days from admission to opioid start and in the initial and final doses of opioids between the 2 groups, the duration of opioid use was significantly longer in the oral group (12.5 vs. 2.0 days; P<0.001). All patients in the i.v./s.c. group died during hospitalization, with a median duration of use of 2 days and a maximum of 10 days. In contrast, 6 (33%) patients in the oral group were discharged alive, with a median duration of use of 12.5 days and a maximum of 661 days (P=0.03). The Kaplan-Meier curve showed that the survival rate was significantly higher in the oral than i.v./s.c. group (P<0.0001; Figure 2). Although side effects were more common in the oral group, there was no significant difference in between the 2 groups (39% vs. 8%; P=0.09).
| Oral group (n=18) |
i.v./s.c. group (n=13) |
P value | |
|---|---|---|---|
| Time from admission to opioid start (days) | 10.6±9.5 | 10.7±9.3 | 0.98 |
| Initial dose (mg/day) | 2.5 [2.5–5] | 5.8 [2.5–10] | 0.1 |
| Medication at initiation | |||
| Codeine | 9 (50) | ||
| Hydromorphone | 9 (50) | ||
| Intravenous morphine | 11 (85) | ||
| Subcutaneous morphine | 2 (15) | ||
| Final dose (mg/day) | 5.0 [2.5–8.75] | 7.5 [2.5–10.0] | 0.34 |
| Duration of use (days) | 12.5 [4.5–29.0] | 2.0 [2.0–4.0] | <0.001 |
| Side effects | 7 (39) | 1 (8) | 0.09 |
| Drowsiness | 1 (6) | 1 (8) | |
| Constipation | 6 (33) | 0 (0) | |
| Nausea | 1 (6) | 0 (0) | |
| Survival discharge | 6 (33) | 0 (0) | 0.03 |
Unless indicated otherwise, values are expressed as mean±SD, median [interquartile range], or n (%). Opioid doses are expressed as morphine equivalents.
Kaplan-Meier curve for all-cause mortality in patients in whom opioids were initiated intravenously or subcutaneously (i.v./s.c. group) and those in whom they were initiated orally (oral group).
Serial changes in vital signs and STAS-J score at the start of opioid administration and 24 and 72 h later are shown in Figure 3. The i.v./s.c. group showed significantly greater changes in vital signs, except for oxygen saturation, whereas the oral group generally showed no change. The STAS-J scores, which assessed dyspnea, improved significantly in both groups (P<0.0001), with no difference between them.
Serial changes in (A) heart rate, (B) systolic blood pressure, (C) respiratory rate, (D) oxygen saturation, and (E) and the Japanese version of the Support Team Assessment Schedule (STAS-J) score in patients in whom opioids were initiated intravenously or subcutaneously (i.v./s.c. group) and those in whom they were initiated orally (oral group).
Palliative care should be provided to patients with advanced HF. However, there are compounding barriers to palliative care in the field of HF, such as difficulties in predicting and recognizing the end stage of the condition, a lack of established medical treatment regimens that include opioids, a lack of insurance approval of drugs, and misconceptions about palliative care such as that it is only for patients with cancer or for the last weeks of life.10–12 Against this background, we have introduced a protocol for oral opioid use for dyspnea in patients with advanced HF at Saitama Citizens Medical Center, and this retrospective study has investigated our experience with protocol-driven oral opioid use. The results of this study indicate that low-dose oral opioids could alleviate the dyspneic symptoms due to HF without significant changes in vital signs, with a survival rate that was significantly higher than for the i.v./s.c. group. No serious adverse effects, including respiratory depression, were observed with oral opioid medications. Thus, we believe that oral opioids may be practical and effective for dyspnea in patients with advanced HF.
Route of Opioid AdministrationAccording to World Health Organization guidelines for the pharmacological and radiotherapeutic management of cancer pain in adults and adolescents, analgesic medicine should be administered “by mouth”, “by the clock”, “for the individual” and with “attention to detail”.13 Current guidelines for chronic HF in Japan also recommend low-dose, sustained-release oral morphine for refractory dyspnea.7 However, in clinical practice, morphine is often initiated i.v. or s.c. in patients with HF.12 In the study of Kawaguchi et al,14 26 of 43 patients (60%) initially received morphine via i.v. or s.c. injection. Similarly, in the present study, morphine was initially administered by i.v. or s.c. injection in 13 of 31 patients (42%), even after the introduction of our protocol. Considering that oral administration of morphine was difficult in these 13 patients, hence the selection of i.v. or s.c. injection, may indicate that these patients were in a poor general condition and that the introduction of an opioid was delayed. Conversely, the oral group included more patients who had been hospitalized within 1 year or who were receiving treatment such as tolvaptan or continuous infusion of dobutamine, suggesting repeated consultations and established relationships between medical staff and patients. Although the present study cannot determine the factors contributing to the difference in survival rate between the 2 groups, it is possible that the timing of the introduction of opioids during the progression of HF may have had an effect. To introduce palliative care at the appropriate time in patients with advanced HF, we think it is desirable to establish rapport and adequate consultation between the patient and medical staff.
Selection of Oral OpioidsIn Japan, several oral opioids, such as codeine, morphine, tramadol, oxycodone, and hydromorphone, are available for patients with cancer, whereas only morphine is approved and recommended for dyspnea in HF patients.7,15 Because the metabolites of morphine are excreted by the kidneys, the dosage of morphine preparations must be adjusted in patients with renal dysfunction. In particular, many patients with advanced HF have renal dysfunction and require dosage adjustments, but oral morphine needs adjustment of a powder form and it needs to be administered orally multiple times a day due to its short duration of action.
Codeine is also approved and used for severe cough, pain, and dyspnea in patients with cancer and respiratory disease. In the protocol reported on here, codeine was adopted as the first-line oral opioid. Overall, 9 of the 18 patients in the oral group initially received codeine and 5 patients continued this drug. Although codeine provided symptomatic relief in only a few patients, it is considered an opioid option for dyspnea in patients with advanced HF.
Hydromorphone used in our protocol is a semisynthetic opioid that is available in Japan for pain and dyspnea in patients with cancer. Hydromorphone is not metabolized by the cytochrome P450 enzyme pathway, thus reducing the potential for significant drug-drug interactions. In addition, hydromorphone is metabolized in the liver by glucuronidation into hydromorphone-3-glucuronide, which has weak pharmacological activity; this makes it less susceptible to genetic diversity, drug interactions, and renal dysfunction. Hydromorphone can be initiated at a low dose and, in extended-release form, its effect can be maintained when taken once a day. In addition, immediate-release hydromorphone can be used as a rescue medication for sudden dyspnea.
Oxycodone is also a semisynthetic opioid that is indicated for moderate to severe pain and dyspnea in cancer patients. Tanaka et al reported a case in which oxycodone was effective in treating dyspnea in a patient with end-stage HF and renal insufficiency.16 Although oxycodone is not used in our protocol, it is expected to be an option for palliative care for dyspnea in patients with advanced HF, similar to hydromorphone.
Effects and Side Effects of OpioidsAlmost 90% of advanced HF patients have been reported to experience dyspnea with minimal exertion or at rest, which substantially limits their ability to perform daily activities.5 Low-dose opioids are reported to be effective for dyspnea in patients with cancer and other diseases.17 The mechanisms by which opioids relieve dyspnea include decreasing the respiratory rate, tidal volume, central perception of dyspnea, and anxiety, as well as altering peripheral opioid receptor activity in the respiratory tract.17,18 Although there is insufficient evidence regarding the appropriate use of opioids in advanced HF, it is known that low doses of opioids can alleviate dyspnea compared with doses used for pain in cancer patients. In our study, morphine-equivalent doses in the oral group were low, with an initial dose of 2.5 mg/day (median) and a final dose of 5.0 mg/day (median). In 16 of 18 evaluable patients in the oral group, the dyspnea score, as determined by STAS-J, improved with low-dose oral opioids. In a study by Kawaguchi et al, the dose of morphine was 5–10 mg/day, which is thought to reflect the real-world clinical practice.14 Because these were single-center retrospective studies with a small sample size and no placebo comparison, the results obtained cannot suggest significant efficacy, but the symptom scores for dyspnea improved in patients with advanced HF. Meanwhile, previous clinical randomized double-blind studies failed to establish the short- and long-term efficacy of oral morphine for HF patients.19,20 These 2 studies used slightly higher doses of morphine (10–20 mg/day), and the majority of patients were in New York Heart Association (NYHA) functional class III, which does not correspond to the group administered morphine in Japan (most patients are NYHA Class IV). Although the routine use of opioids for dyspnea is not recommended by the 2021 European Society of Cardiology guidelines,21 a meta-analysis on their effects for chronic HF patients showed an advantage in exercise testing, but most studies included NYHA Class II/III patients and morphine use ranged from 10 to 20 mg/day.22 Regarding palliative care for HF patients, there are differences in the background of opioid use in Japan and clinical results in other countries, so the initiation of opioid use requires thoughtful consideration of patient selection, opioid type, dosage, and assessment of efficacy and side effects.
In the oral group in our study, 8 of 18 patients (44%) experienced side effects, but there were no serious side effects, such as respiratory depression, disturbance of consciousness, hypotension, or bradycardia. The majority of side effects were constipation in patients receiving oral codeine, which could be managed with laxatives. Johnson et al reported that morphine-related adverse events are generally mild and include constipation, nausea, and vomiting, the latter 2 of which are mostly self-limiting.17 A meta-analysis found no evidence of significant or relevant adverse respiratory effects of opioids for chronic breathlessness.23 However, patients with advanced HF are mostly elderly, which means that they may have impaired renal or liver function, and side effects should be carefully assessed because of drug accumulation with long-term use.
Study LimitationsA major limitation of this study is that it was a single-center retrospective investigation with a small number of patients. The subjects were admitted to Saitama Citizens Medical Center for HF, and their average age was >80 years. In addition, there was a risk of patient selection bias because the decision to apply this protocol and when to use it depended on the judgment of the attending physician. It is thus difficult to apply the results of this study to all patients with advanced HF. A second limitation of the study is the ambiguity in symptom assessment. In this study, STAS-J was used to assess dyspnea, but such an evaluation may be difficult in some cases. Similarly, evaluation regarding side effects is sometimes difficult, especially in patients close to the end of life. For example, three-quarters of patients in the i.v./s.c. group died within 72 h of initiating morphine, making it difficult to determine whether the death was due to disease progression or drug use. In the unstable condition of advanced HF, it is difficult to accurately assess the effects and side effects of opioids. Third, we only used a limited range of opioids in this study, and their differences from other opioids are unclear.
We have introduced a protocol on opioid use for dyspnea in patients with advanced HF and retrospectively reviewed its use. This single-center study suggests that oral low-dose opioids may be practical and effective. However, there are issues that need to be resolved regarding the use of opioids, such as their coverage under Japanese national health insurance, and further studies are warranted to examine efficacy and safety. In addition to appropriate medical treatment, the need for assessment and intervention regarding living conditions, social background, nutrition, physical activity, and psychology is also emphasized.
None.
This research received no grant from any agency in the public, commercial, or not-for-profit sectors.
The authors declare that they do not have any conflicts of interest to disclose.
The study was approved by the Institutional Review Board for Human Investigation of Saitama Citizens Medical Center (Approval no. 2019-26). It also conformed to the principles outlined in the Declaration of Helsinki.
The deidentified participant data will not be shared.
