Clinical practice guidelines use and general information

One challenge to developing protocols and clinical practice guidelines in paediatric cardiology is the significant heterogeneity of disease. When addressing postoperative pain management, the wide variety of underlying disease and various surgical options challenges the ability to decrease practice variability. Collaboratives offer the opportunity to amass multi-centre data and experiences, thereby increasing understanding of a given topic, such as postoperative pain management. ^{Reference Thompson, Foote and King19} While clinical practice guidelines are helpful in providing guidance to clinicians, it is important to recognise that they cannot address all patient populations and judgement must be used when applying them. Individual patients and high-risk populations may not benefit from application of clinical practice guidelines. ^{Reference Fuller, Kumar and Roy3} Involvement of all stakeholders and participants in clinical care is important when developing protocols. Bedside nurses, in particular, are well positioned to improve associated processes as they frequently assess and treat pain. ^{Reference Bastero, DiNardo, Pratap, Schwartz and Sivarajan21–Reference Magner, Valkenburg and Doherty24}

Pain assessment

Undertreatment or reluctance to treat pain among children with cardiac disease is likely common and often related to concerns that analgesics may have unfavourable haemodynamic effects. However, undertreatment may negatively result in increased stress response which may have negative long- and short-term consequences. ^{Reference Diaz6,Reference Hehir, Easley and Byrnes7} A variety of tools have been developed to assess pain in infants and children, utilising both subjective and objective information. There is insufficient evidence to suggest that any of these tools is superior, however consistent use and unit-wide acceptance of a given tool improves consistency of pain assessment and subsequent treatment. ^{Reference Diaz6,Reference Hehir, Easley and Byrnes7,Reference Franck, Ridout, Howard, Peters and Honour25–Reference Bai, Hsu, Tang and Van Dijk27} Pain assessment among infants and children, particularly those who are non-verbal, relies on a combination of factors including objective data such as vital sign changes and subjective information such as patient-reported symptoms or family assessment. ^{Reference Huth, Broome, Mussatto and Morgan29} There are currently efforts to develop additional tools to objectively measure pain; however, these efforts have been primarily limited to adult populations. ^{Reference Pollak, Bronicki, Achuff and Checchia5} Children would benefit from such tools, especially non-verbal children or children who are developmentally unable to express their pain. It is important to recognise that social and cultural factors may influence the experience of pain and treatment should be adjusted based on individual response. ^{Reference Pollak, Bronicki, Achuff and Checchia5,Reference Diaz6,Reference Bai and Hsu28} Though we did not identify literature examining the impact of implicit bias or racism on the treatment of pain among children undergoing congenital heart surgery, this is a known problem in medicine ^{Reference Sabin and Greenwald30–Reference Fanta, Ladzekpo and Unaka32} and pain should never be managed differentially based solely on a patient’s race or ethnicity.

General principles

Pain in the postoperative period should be considered in the context of the individual patient and their risk factors for ongoing pain. Specifically, the incision type and presence of chest tubes may predispose patients to ongoing pain, requiring additional analgesia. ^{Reference Karuppiah, Pehora, Haller and Taylor33} Additionally, timing within a hospitalisation should be taken into account given the temporally associated reduction of pain as patients' progress from their surgery. As patients’ pain improves and their need for analgesia diminishes, it is most appropriate to first discontinue or reduce frequency of administration of medications with the highest risk for side effects and continue those medications which have fewer side effects. It is appropriate to utilise additional medications to treat known and common side effects of commonly used analgesics. In particular, a bowel regimen should be implemented for patients taking opioid medications as opioids are associated with constipation.

Pain that does not follow the anticipated course over time, or that is inconsistent in type or location, or is associated with other signs and symptoms such as vital sign changes or respiratory distress is cause for rapid and thorough evaluation for other causes. Similarly, additional evaluation is needed when patient or family reports of pain are discrepant from provider’s assessment of pain.

Preoperative considerations

Families and providers often have discrepant expectations for postoperative pain level and management which can challenge the therapeutic relationship needed to assist children in their recovery. Preoperative counselling should therefore address realistic expectations for postoperative pain and aim to establish understanding with families regarding what their child may experience and how it will be addressed. ^{Reference Huth, Broome, Mussatto and Morgan29,Reference Harvey and Kovalesky34} Discussions related to goals of pain management should emphasise functionality rather than complete elimination of pain. Additionally, management of stress and anxiety in the preoperative setting has been associated with reduced postoperative pain. ^{Reference Kumar, Das, Chauhan, Kiran and Satapathy35}

Intraoperative considerations

It has been suggested that multimodal pain management during surgery, including use of acetaminophen, local anaesthesia and postoperative nausea and vomiting treatments, may reduce the need for intraoperative opioids and improve postoperative pain outcomes. ^{Reference Roy, Parra and Brown4} There have been a number of studies examining the use of intraoperative methadone and its impact on postoperative pain but based on the current evidence, this panel does not currently have a recommendation related to its use. ^{Reference Iguidbashian, Chang, Iguidbashian, Lines and Maxwell36–Reference Robinson, Caruso, Wu, Kleiman and Kwiatkowski38}

Regional anaesthesia

The use of regional anaesthesia has gained interest and acceptance as a means of reducing intraoperative and postoperative opioid requirements. Currently, centre-specific considerations are required for use of regional anaesthesia-based experience of anaesthesiologists and availability of providers to manage such modalities outside of the operating room such as consultative pain services. Multiple studies have examined specific types of regional anaesthesia in the context of paediatric congenital heart surgery and it has widely been found to be safe. Specifically for transversus thoracis plane blocks, studies have shown improved pain scores, decreased intraoperative and postoperative opioid requirements, shorter time to extubation and shorter ICU length of stay compared to patients not receiving nerve blocks. ^{Reference Abdelbaser and Mageed39–Reference Cakmak and Isik41} Similar endpoints have been used in studies of paravertebral block and have shown promise though the body of evidence is insufficient currently to uniformly recommend paravertebral block. ^{Reference Türköz, Balcı and Can Güner42,Reference Sahajanandan, Varsha and Kumar43} Additionally, erector spinae plane blocks and parasternal intercostal blocks using continuous infusions shown to be safe but there is inconclusive evidence to suggest uniform adoption of such approaches. ^{Reference Macaire, Ho and Nguyen45–Reference Kaushal, Chauhan and Saini51} No specific anaesthetic agent has shown superiority in these studies. Caudal analgesia has been shown to be safe with minimal harm. However, there is controversy regarding its impact on postoperative pain as studies comparing use of caudal analgesia to children not receiving regional anaesthesia have not consistently demonstrated improved outcomes; therefore it is reasonable to consider caudal analgesia but as it does not have clear benefits, it is not uniformly recommended. ^{Reference Samantaray, Trehan, Chowdhry and Reedy52–Reference Maharramova and Taylor56} Spinal anaesthesia may improve pain scores and reduce postoperative opioid requirement however only one study with a small sample size of 20 patients has demonstrated this. ^{Reference Hammer, Ramamoorthy and Cao57} While many forms of regional anaesthesia have conflicting evidence, continuous infusion of local anaesthetic at incision sites has been shown to reduce postoperative pain scores and opioid requirements and should be considered, though some evidence to support this comes from adult studies. ^{Reference Nasr and Abdelhamid54,Reference Tirotta, Munro and Salvaggio58,Reference Paladini, Di Carlo and Musella59} Further research is needed to explore the efficacy of these various regional anaesthetic techniques in order to identify which patients might benefit from regional anaesthesia.

Opioids

Despite side effects and risk for habituation, opioids remain a mainstay of postoperative pain management and offer effective analgesia. ^{Reference Chou, Gordon and de Leon-Casasola69} There are some specific considerations regarding opioid use among children undergoing cardiothoracic surgery. In animal studies, chronic hypoxaemia has been associated with increased sensitivity to opioid medications. Studies among humans with obstructive sleep apnoea and chronic hypoxaemia have shown similar results prompting consideration of whether children with cyanotic CHD may also have variable sensitivity to opioids. When studied, however, children with cyanotic lesions were not found to have differing sensitivity to opioids and therefore should be prescribed typical doses. ^{Reference Murray-Torres, Tobias and Winch61}

Another specific consideration relates to children with Trisomy 21 and CHD. Previous reports have suggested possible altered pharmacokinetics or pharmacodynamics among children with Trisomy 21; however, multiple studies have found no difference among children with and without Trisomy 21, and these children should therefore receive opioids at typical doses. ^{Reference Goot, Kaufman and Pan64–Reference Van Driest, Shah and Marshall67}

There have been limited published reports related to decreased opioid clearance among individuals with decreased systemic blood flow prompting consideration for need to dose opioids differentially among individuals with CHD. ^{Reference Elkomy, Drover and Glotzbach68,Reference Dagan, Klein, Bohn, Barker and Koren70} Despite these reports, modelling of pharmacokinetics has shown that children with CHD do not metabolise morphine significantly differently from children without CHD and should therefore be treated with conventional doses. ^{Reference Elkomy, Drover and Glotzbach68} Patients with renal dysfunction, a common issue following cardiac surgery, are at risk for toxic accumulation of metabolites of morphine. ^{Reference Elkomy, Drover and Glotzbach68}

Among certain populations, patient-controlled analgesias have been shown to have increased effectiveness of analgesia compared to provider-initiated intermittent boluses. ^{Reference Chou, Gordon and de Leon-Casasola69} Limited studies among children undergoing cardiac surgery have shown similar findings. ^{Reference Mota, Marcolan, Pereira, Milanez, Dallan and Diccini62,Reference Epstein63} Because of the potential for better analgesia, higher satisfaction, and overall reduction in opioid exposure, patient-controlled analgesias should be considered for children who require intravenous opioids and are developmentally appropriate.

Opioid-sparing

Due to increased awareness of risks associated with opioid medications including habituation, withdrawal, and side effects including constipation and respiratory depression, as well as evidence that opioid-sparing approaches may facilitate postoperative recovery, there is significant interest in developing protocols and approaches to postoperative pain management which minimise opioid exposure. ^{Reference Chou, Gordon and de Leon-Casasola69,Reference Frankel, Maul and Chrysostomou71} Multimodal pain management has been utilised successfully to reduce opioid exposure postoperatively. ^{Reference Roy, Parra and Brown4} Additionally, increased awareness of the likely trajectory of patients in the immediate postoperative period, particularly those who are likely to tolerate extubation soon after arrival in the ICU, has facilitated opioid-sparing approaches. ^{Reference Frankel, Maul and Chrysostomou71} In a randomised controlled trial of children recovering from cardiothoracic surgery, there was no difference in pain control among those who received continuous versus intermittent opioids but use of continuous infusions resulted higher total doses of opioids and longer length of stay. ^{Reference Penk, Lefaiver, Brady, Steffensen and Wittmayer72} Increased examination of whether continuous infusions are needed for individual patients may result in decreased reflexive use of them and improved patient outcomes.

Non-opioid medications

Multimodal pain management relies largely, though not exclusively, upon NSAIDs and acetaminophen. While NSAIDs are used in neonates and infants for indications such as closure of the patent ductus arteriosus, retinopathy of prematurity, and prevention of intraventricular haemorrhage, there has been reluctance to use NSAIDs as a primary means of analgesia among infants less than 6 months of age. ^{Reference Aranda, Salomone, Valencia and Beharry89,Reference Ziesenitz, Zutter, Erb and van den Anker90} Additionally, there is significant variation in practice related to NSAID and acetaminophen or paracetamol use in children. A systematic review of guidelines for anti-pyresis among children identified 55 guidelines addressing use of ibuprofen and 69 addressing use of paracetamol. There was minimal agreement between guidelines regarding appropriate population, dosing, frequency, duration of treatment, or conditions in which to avoid or use caution prior to initiating the drug. ^{Reference Green, Krafft, Guyatt and Martin91} Additionally, for many of the age-based recommendations against use of NSAIDs or acetaminophen in young children, there was either no published evidence or low quality of evidence to support those recommendations. ^{Reference Green, Krafft, Guyatt and Martin91} More recent publications have examined use of NSAIDs in younger children and found them to be safe. ^{Reference Ziesenitz, Zutter, Erb and van den Anker90}

Multiple studies have examined use of ibuprofen or ketorolac in postoperative infants and children, some as young as 28 days. In addition to focusing on achieving analgesia, these studies have largely used creatinine or blood urea nitrogen as endpoints. ^{Reference Saini, Maher and Deshpande74–Reference Dawkins, Barclay, Gardiner and Krawczeski76} These are known to be imperfect markers of true renal function and, while they were found to be increased among patients receiving NSAIDs, the clinical implication of that rise is unclear, particularly as the values did not always fall outside of the range of normal. Acute kidney injury following cardiac surgery is common but is multifactorial in aetiology and not directly related to nephrotoxin exposure, including ibuprofen and ketorolac. ^{Reference Uber, Montez-Rath, Kwiatkowski, Krawczeski and Sutherland92} Additionally, studies examining the risk of bleeding and changes to platelet function while receiving NSAIDs have found platelet function is altered but there is no increase in clinically significant bleeding. ^{Reference Savva, Kishk, Morgan, Biggs, Seung and Bauer73–Reference Dawkins, Barclay, Gardiner and Krawczeski76,Reference Gupta, Daggett, Ludwick, Wells and Lewis79,Reference Kim, Kaufman, Patel, Manco-Johnson, Di Paola and da Cruz88} It is unclear what the implication of changed platelet function is, and this warrants further research, particularly considering the additional risk factors in this population including cardiopulmonary bypass, presence of chest tubes, and surgical wounds.

In addition to NSAIDs and acetaminophen, there have been a few studies examining use of other non-opioid analgesics among adults and there have been suggestions to study these among children. Perioperative administration of magnesium has been shown to decrease postoperative pain and opioid consumption among adults. ^{Reference Pollak, Bronicki, Achuff and Checchia5} While there are no data, to our knowledge, among children, it is a worthwhile consideration given the relative ease of monitoring magnesium levels (something that is often already occurring in postoperative patients) and the low risk of side effects when administered enterally or slowly if parenterally. Additionally, gabapentinoids have been identified as potentially useful in children recovering from cardiac surgery. Pregabalin, in particular, appears to be effective for acute pain and has been studied among adults recovering from cardiac surgery. ^{Reference Pollak, Bronicki, Achuff and Checchia5} We could not identify any data related to spasmolytic medications but in practice, these are commonly utilised after CHD surgery, particularly in adolescents and young adults. Further research is warranted to explore these, and other, non-opioid medications for treatment of pain in children.

Non-pharmaceutical pain management

Non-pharmacologic approaches to analgesia are widely accepted as an important component of postoperative care, but may be underutilised among children undergoing and recovering from cardiac surgery. ^{Reference Chou, Gordon and de Leon-Casasola69,Reference Miller, Lisanti and Witte93} Reasons for underutilisation include lack of resources and inadequate awareness of benefits but services such as music therapy, massage therapy, acupuncture, and others have been shown to improve postoperative pain scores in this population. ^{Reference Miller, Lisanti and Witte93–Reference Huang, Lei, Liu, Cao, Yu and Chen95,Reference Kakar, Billar, van Rosmalen, Klimek, Takkenberg and Jeekel97–Reference Albert, Gillinov, Lytle, Feng, Cwynar and Blackstone100,Reference Mayan, Alvadj, Punja, Jou, Wildgen and Vohra103} Additionally, not all non-pharmacologic approaches are resource intensive – indeed, skin-to-skin contact has been shown to improve comfort among infants undergoing cardiac surgery and increase physiologic stability. ^{Reference Lisanti, Demianczyk and Costarino104} Other interventions as simple as providing developmentally appropriate toys and optimising environmental factors such as lighting and noise can also improve postoperative outcomes related to pain. ^{Reference Diaz6,Reference Chen, Lei, Liu, Cao, Yu and Chen102} Other non-pharmacologic analgesia modalities have been shown to be effective among adults undergoing thoracic surgery, chiefly transcutaneous electrical nerve stimulation. ^{Reference Chou, Gordon and de Leon-Casasola69} Transcutaneous electrical nerve stimulation has also been shown to be effective in reducing procedural pain for children. ^{Reference Lander and Fowler-Kerry105} Efforts should be made to determine if transcutaneous electrical nerve stimulation is effective among children undergoing cardiac surgery, particularly considering the lack of side effects and ease of use. Importantly, there is minimal risk associated with these non-pharmacologic modalities.

Discharge considerations

Children are likely to need ongoing analgesia after discharge home following cardiac surgery. Discharge teaching may even include consideration that pain can transiently worsen as children become more mobile in their home environments. Acetaminophen and NSAIDs should be used as the first-line medications for treatment of pain in the home environment – children still requiring frequent administration of opioids may not yet be ready for discharge. ^{Reference Frankel, Maul and Chrysostomou71} To that end, Monitto et al studied the effectiveness of analgesia following discharge for a variety of paediatric surgeries including cardiac surgery and found that despite low levels of use of opioids at home, analgesia was largely good or excellent at 1–2 days following discharge and 10–14 days following discharge. ^{Reference Monitto, Hsu and Gao107} Because so few patients require frequent opioid administration at home and the majority of prescribed doses are unused, it is reasonable to limit the number of prescribed doses and some patients may not require opioids at all. ^{Reference Irfan, Martin and Canner106,Reference Monitto, Hsu and Gao107} In this study, atrial septal defects made up the majority of included patients making it easiest to apply their findings clinically to that population. However, it should be noted that much of the postoperative pain experienced by patients is related to incision sites (sternotomy, thoracotomy, chest tubes when in place) rather than the heart itself and therefore the specific form of CHD likely has little impact on postoperative pain and rather the extent of the incision, first time versus redo incision, incision healing, and features such as delayed sternal closure play heavily into the experience of pain.

Studies of opioid prescription at discharge among adults undergoing cardiac surgery have shown that the greatest predictor of amount of opioid prescribed is hospital, not patient or surgery-specific factors. ^{Reference Holst, Dearani and Schaff108} There remains a need for research to better understand best practices surrounding opioid prescription at time of discharge for children undergoing cardiac surgery, particularly given the risks of undertreating pain in the home environment relative to the risks of overuse or misuse of opioid medications. Studies targeting opioid prescription variability for postoperative children have shown that institutional guidelines available to prescribing physicians can reduce variability and decrease overall opioid prescription at time of discharge, though these results have not been reproduced among children undergoing cardiac surgery. ^{Reference Irfan, Martin and Canner106}