Worldwide, cancer is diagnosed in approximately 300,000 children under age 20 years each year, and childhood cancer contributes to 80,000 deaths per year.¹ In the United States, almost 16,000 cases of cancer are diagnosed each year in children aged birth to 19 years, contributing to almost 2,000 deaths per year.² As recently as the 1970s, a diagnosis of cancer during childhood was considered a uniformly fatal disease. However, through enrollment in clinical trials available through cooperative pediatric oncology groups, as well as advances in supportive care, 5-year overall survival rates in childhood cancer are now over 80% in the United States.³ The 12 major diagnostic categories of childhood cancer in The International Classification of Childhood Cancers, third edition (ICCC-3) include (1) leukemias and myeloproliferative and myelodysplastic diseases, (2) lymphomas, (3) central nervous system (CNS) neoplasms, (4) neuroblastoma and other peripheral nervous cell tumors, (5) retinoblastoma, (6) renal tumors, (7) hepatic tumors, (8) malignant bone tumors, (9) soft tissue and other extraosseous sarcomas (e.g., rhabdomyosarcoma), (10) germ cell tumors, (11) other malignant epithelial neoplasms and malignant melanomas (e.g., thyroid and nasopharyngeal carcinomas), and (12) other unspecified malignant neoplasms.⁴ Certain types of cancer are more common in specific age groups of children. For example, acute leukemias, neuroblastoma, and renal tumors are more common in the younger age groups (ages 2 to 6 years), whereas the malignant bone tumors and lymphomas are more common during the adolescent years.

The end result of these aggressive treatment regimens for pediatric cancer patients is a success story for the 21st century. The current estimated survival rate for a pediatric cancer patient is 80%. The current number of survivors within the United States is over 370,000.^2,3 However, despite these advances, there continues to be a high risk for the development of chronic health conditions (late effects) years after the cancer treatment is completed. Complications of treatment include risk for second malignant neoplasms, cardiac and/or pulmonary dysfunction, skeletal problems, and neurocognitive deficits.⁶ There is nearly a 75% risk of childhood cancer patients having a chronic health problem by 30 years postdiagnosis.⁷ As the population of childhood cancer survivors continues to grow, it is important to be familiar with the unique challenges faced both during and after completion of treatment for cancer during childhood.

From birth through early childhood, a normal developmental assessment evaluates five main areas: gross motor skills, fine motor skills, language skills, personal/social skills, and cognitive skills. Changes occurring in these areas impact the pain assessment and emotional response of the child to the painful stimuli. For example, the language skills of a 2-year-old include a 50-word vocabulary and 2-word sentences. During this period, if the child is unable to effectively communicate his or her pain sensation and his or her pain is inadequately treated, there can be more fear and anxiety with each subsequent painful procedure. One year later, at 3 years of age, there is an expected 250-word vocabulary, 3-word sentences, and speech is intelligible to strangers 75% of the time. This child may be able to more effectively communicate with his or her parents and doctors and have a treatment for the pain initiated more promptly, a factor related to these enhanced communication skills that can serve to decrease anxiety for future procedures.

A school-age child will have a progressive ability to effectively communicate with the health care team. In turn, the providers must clearly communicate with the child about their treatment
plan in order to minimize the anxiety children can experience from medical interventions. During adolescence, normal development of increasing desire for autonomy necessitates direct communication by the health care team, not only with parents but also directly with the adolescent based on the adolescent’s desire for independent decision making (when possible). However, in all levels of normal development, the impact of illness can cause a pediatric patient to regress and become more dependent on his or her parents to support physical and emotional needs, with parents often providing the primary input to the health care team about children’s pain and effectiveness of pain management. When possible, it is always helpful for the health care team to also attempt to get the child or adolescent’s self-report of pain.

Pain is a prevalent symptom accompanying pediatric cancer. Many patients (49% to 62%) with pediatric cancer have pain as a presenting symptom at diagnosis, with many of these patients having pain as the sole presenting symptom of their cancer.^9,10,11 Pain is often present for several months prior to diagnosis.⁹ Using surveys during treatment, Miser et al.¹² found nonprocedural pain was prevalent for 54% of inpatients and 26% of outpatients in a single center. A national study in Germany by Zernikow et al.¹³ found 15%, 28%, 50%, and 58% of patients diagnosed with cancer to be in pain at time of, within 24 hours, within 7 days, or within 28 days of interview, respectively. This is corroborated by Fortier et al.¹⁴ who used a daily diary to show that 53.3% of children diagnosed with cancer had chronic or recurrent pain, with 40% of children requiring at least one dose of analgesic over the 14-day study period. Pain is more prevalent in pediatric inpatients with cancer versus outpatients (84% vs. 35%) and most prevalent in children with solid tumors (63%) versus leukemia (44%), lymphoma (27%), or CNS tumors (50%).¹⁵

In contrast to findings in adults, pain in children undergoing treatment for cancer was predominantly caused by antineoplastic treatments and procedures, rather than from the disease itself.^{10,11,13,16,17,18} In one study, 49% of children stated that treatment-related pain was the most severe problem, followed by 38% of children for procedure-related pain, and finally 13% for disease-related pain.¹⁰ Disease-related pain generally becomes less prevalent over time.^9,11,12 The difference in etiology between adult and pediatric cancer pain reflects differing cancer diagnoses and treatment strategies for adult and pediatric cancers. Although pediatric cancers are often rapidly proliferating mesenchymal tumors or leukemia and thus cause pain at diagnosis, many are also responsive to an aggressive combination of intravenous and intrathecal chemotherapeutic agents, surgery, and/or radiation, which all can lead to painful therapy-related pain. In contrast, adult cancers are often slow-growing epithelial tumors which are often treated with less aggressive therapies.¹³

Increased pain prevalence is reported for certain groups of patients. Among children with acute lymphoblastic leukemia (ALL), pain is reported by the majority of patients throughout the first year of treatment, improving somewhat over time.^19,20 In children with progressive, recurrent, or nonresponsive cancer, pain prevalence is 49%, rising to 62% in the last 12 weeks of life. Furthermore, high-distress pain prevalence in the same group of children is 39%, rising to 58% in the last 12 weeks of life.²¹ By the last month of life, other studies show pain prevalence at 82%, with about half of all patients suffering “a great deal” or “a lot” from the pain.²²

Undertreatment of pain continues to be a concern in pediatric cancer, both in the inpatient setting^23,24,25 and in the outpatient and home settings.^14,19 Pain in these settings continues to be common, underrecognized, and undertreated. At the end of life, treatment of pain is often unsuccessful.²²

Barriers to adequate pain control in pediatrics include lack of pain documentation and assessment^18,25,26 and clinical staff reluctance to prescribe or administer analgesics²⁵ whether because of fear of addiction²⁷ or inadequate pain management knowledge.^27,28 Some clinicians hold the mistaken belief that infants and children are less sensitive to pain.²⁷ Physicians have been found to underestimate children’s pain as reported by the patient or the parent.²⁹ Other common misconceptions include beliefs that a significant percentage of children overreport pain, that increases in morphine dose beyond a certain amount will not produce increased pain relief, and that respiratory depression is likely to happen in patients who have received opioids for months.^30,31,32 Breakthrough pain, defined as episodes of medium to severe pain on a background of otherwise controlled pain, can occur suddenly. It can last seconds to minutes, often too fast for many “as-needed” medications to work effectively.²³

In the home setting, barriers to adequate analgesia may include caregiver misconceptions regarding children’s pain, fears of side effects of analgesic use, as well as avoidance of analgesic use.^33,34 Two studies in different cultural contexts have found that many parents reported believing that a child always tells his or her parent when he or she is in pain.^32,35

In adolescents, patient attitudes may also act as barriers to effective treatment of pain. Similarly to adults, adolescents are concerned about addiction, development of tolerance, and analgesic side effects as well as losing their ability to monitor healthrelated bodily changes. Some also hold the fatalistic belief that cancer pain is unavoidable. Some attitudes are more uniquely seen in this age group and stem from developmental needs for autonomy and control, including concerns about how reporting pain may lead to undesirable tests and restrictions in social activities. They are also concerned about not being involved in treatment decisions. Thus, active listening is important when reviewing adolescents’ pain symptoms.^36,37

Finally, system-based barriers exist, including limits to access to opioids and accessibility of pain and palliative care specialists, especially for pediatric populations and resource-poor areas.^38,39,40

Pain, especially from medical procedures, leads to fear and anxiety.⁴¹ Younger children receiving medications to control pain prior to a procedure experienced not only less pain with each procedure but also had consistently lower pain scores for subsequent medical procedures.⁴² Procedure-related pain
contributes to posttraumatic stress symptoms in survivorship.⁴³ Pain with moderate to severe distress was associated with a significant worsening in total health-related quality of life (HRQOL) scores as well as scores in specific domains related to physical, emotional, and school HRQOL.⁴⁴ Adolescents with cancer, especially those experiencing severe pain, were significantly hindered in pursuing their personal goals, which are important to adolescent needs and their normal developmental trajectories.⁴⁵

The importance of eliciting and listening to all details of the pain narrative of the pediatric patient with cancer is critical. Again, what is asked of the child and expected of the child is dependent on his or her developmental level. For the school-age child or adolescent, the pain assessment should occur early on, where there is the support of family, and in a nonthreatening environment. At tertiary care centers that care for large numbers of pediatric cancer patients, child life services are available and can serve as another resource to help elucidate a clear description of the pain narrative. For the infant or toddler, the health care provider is dependent on the parents’ narratives of the painful experience because they are the source of safety and communication for the young child.

Across the entire developmental continuum of pediatrics, the physical examination, including close observation of the child and the family dynamic, is critical to provide additional information about physical factors that contribute to the experience of pain. During the history and physical examination, it is also critical for the clinician to recognize and identify psychosocial factors, such as parental fear or anxiety, which can also increase the suffering associated with pain for the pediatric patient with cancer.

The evaluation of pain often begins with assessment of its sensory experience. The PQRST method is a useful method of assessing the dimensions of the pain sensory experience in many children.^49,50 PQRST includes asking about Palliating/provoking factors, Quality descriptors of the pain, Radiation of the pain to other parts of the body, Site/symptoms/severity of the pain, and Timing/triggers. Different characteristics may suggest neuropathic pain versus nociceptive pain and may help narrow down sources of pain or may suggest different treatment modalities.

The most commonly evaluated sensory dimension is pain intensity, and several one-dimensional self-report scales are available. The Pieces of Hurt/Poker Chip Tool,⁵¹ Oucher,⁵² Faces Pain Scale-Revised,⁵³ and the Visual Analogue Scale⁵⁴ have been extensively validated and have been recommended by several systematic reviews^55,56,57 as the best measures for clinical practice and research. These tools have been validated for a wide range of pediatric age groups—Pieces of Hurt for 3 to 18 years, Oucher for 3 to 12 years, Faces Pain Scale-Revised for 4 to 16 years, and the Visual Analogue Scale for at least 7 years and older. For children 8 years and older, the 11-point numerical rating (0 to 10) is recommended, when possible.⁵⁸ Specific one-dimensional self-report tools are also available for pain location^59,60 and to describe the temporal dimension of pain.⁶¹

Observing a patient’s behaviors can be useful when self-report is not available or unreliable. In young children, facial expressions, body posture, cries, and inconsolability are commonly associated with acute pain. Observations of physiologic indices such as heart rate, oxygen saturation, or sweating can also be useful in acute pain and are sometimes used in behavior scales for critically ill patients.

These observations are reflected in observational pain intensity scales which can be employed in lieu or in addition to self-report scales. Patients whose pain might need to be measured with an observational (i.e., behavioral) scale might be too young (e.g., below 4 years of age); too distressed; impaired in their cognitive or communication abilities; very restricted by bandages, surgical tape, mechanical ventilation, or paralyzing drugs; or whose self-report ratings are considered to be exaggerated, minimized, or unrealistic due to cognitive, emotional, or situational factors. A systematic review by von Baeyer and Spagrud⁶² recommends two scales for brief pain events and procedural pain (Face, Legs, Arms, Cry, and Consolability [FLACC]⁶³ and Children’s Hospital of Eastern Ontario Pain Scale [CHEOPS]⁶⁴), one scale for postoperative pain in the hospital (FLACC), one scale for postoperative pain at home (Parents’ Postoperative Pain Measure [PPPM]⁶⁵), and one scale for critical care and ventilator patients (COMFORT Scale⁶⁶).

However, many of these behaviors and observations normalize as the body adapts to persistent pain, and thus, different tools and observations may need to be utilized for persistent pain. Verbal expressions of pain continue to be an important indicator; however, lack of interest in surroundings, assuming pain-relieving postures, slowness and paucity of movement, and wariness at being moved become more prominent with persisting pain in young toddlers.⁶² Recently, an observational scale incorporating the aforementioned observations was developed and validated specifically for persistent cancer pain in 2- to 6-year-olds (Hétero Evaluation Douleur Enfant [HEDEN]).⁶⁷

In adolescents, nonverbal cues to persistent pain can be more subtle. In addition to decreased activity, adolescents with chronic pain have been described as exhibiting irritability and moodiness, social withdrawal, and sometimes uncooperativeness with treatments and medications, especially as a way to rebel and exert some bodily control when in pain.⁵⁰

It is important to include emotional (affective) and cognitive (evaluative) measures when assessing children for pain. Affective measures reflect the emotional experience of pain. Evaluative measures reflect the child’s ability to cope with his or her pain.

Toddlers with cancer who experience pain display specific pain behaviors but can also display psychomotor inertia and anxiety behaviors.^68,69 Anxiety behaviors include tenseness, hostility, crying easily, and wariness at being moved. On the opposite end of pain expression is withdrawal or psychomotor inertia. Behaviors that suggest this include resignation (lack of resistance), withdrawal, lack of expression, lack of interest in surroundings, and slowness and paucity of movement. This behavioral pattern is sometimes labeled “learned helplessness,” “passive coping behavior,” and “sickness behavior, such as fatigue and malaise.” These depressive symptoms can be confused with actual depression, both for the observer as well as for the patient but may improve once pain is addressed.⁶⁹

There are several multidimensional tools that bring together both questions about pain sensory experience as well as affective or evaluative measures. One of these, the Adolescent and Pediatric Pain Tool (APPT),^70,71 itself derived from the McGill Pain Questionnaire, has been validated in children 8 years and over and has been used in several studies of children with cancer.^18,19 It measures pain intensity, location, quality descriptors, temporal descriptors, and the evaluative and affective dimensions of pain. Affective words include words such as awful, frightening, and suffocating. Evaluative words include words such as uncontrollable, horrible, and annoying.

Pain is but one of the symptoms that children experience while going through cancer. Fatigue, nausea/vomiting, poor appetite, and depression and anxiety are some of the most common, besides pain.¹⁵ Symptoms often cluster and contribute toward overall decreases in HRQOL. For example, pain in children with cancer can often lead to decreased functioning and quality of life.^72,73 As physical activity or play activities can be exacerbated by pain from cancer, many patients instinctively avoid activities, a factor that leads to deconditioning and loss of function. This and the presence of pain itself can lead to depressive symptoms. Sleep is also intuitively affected by pain, leading to fatigue.

The National Institutes of Health (NIH)-sponsored Patient-Reported Outcomes Measurement Information System (PROMIS) initiative has developed a pain interference tool, evaluating the presence of psychological, social, and physical dysfunction secondary to pain.⁷⁴ The questionnaire asks patients if pain has hindered their function in regard to sleeping, attention, doing schoolwork, standing, walking, running, being happy, or being angry. Data from PROMIS shows strong positive correlations between pain interference and fatigue and pain interference and depressive symptoms and a strong negative correlation between pain interference and physical functioning-mobility.⁷⁵ Symptom cluster analyses show that pain in adolescents with cancer often co-occurs in the same patients as fatigue, sleep disturbance, and depression⁷⁶ or with nausea and vomiting.⁷⁷ Another cluster analysis study using PROMIS data in children with cancer showed that patients with high pain interference scores are more likely to also have high anxiety, depression, and fatigue scores.⁷⁸

Based on these data, clinicians should assess the effect of the child’s pain on their function in several different domains, including physical functioning and health; mood, coping, and psychological well-being; social and family relationships; and fatigue and sleep.

Treatments for pain along with patient pain relief, functional improvement, and adverse effects should be noted in initial and ongoing evaluations of the patient. Past pain treatments, especially when associated with less than optimal control of pain, should be critically evaluated for appropriate medication or modality, dosing, route of administration, and both beneficial and adverse effects on pain and function. Common reasons for pain treatment failure include intolerable adverse effects, too long of a dosing interval, too small of a dose, or inappropriate route of administration. In addition, not all pain treatments are equally effective for each patient, and thus, it is helpful to know the history of what has been attempted in the past.

Preexisting medical and psychiatric conditions are less common in children but play a role in their ability to cope, physically or psychologically, with cancer treatments. A patient and family’s social situation and spirituality may also play a role in the support systems available in coping with both disease diagnosis and pain and symptom experience.

Whether or not a pediatric patient is able to provide a good history of his or her pain, it can be helpful to elicit observations and history from parents and other caretakers. Often, proxy report is the only source of information when patients are too young developmentally, in distress, or unwilling to provide this history. In addition, pediatric oncology patients are sometimes in the hospital so frequently, or for such an extended period of time, bedside nurses in oncology units and infusion centers will also have a longitudinal perspective on the patients and their current pain. On the other hand, the clinician needs to be mindful that proxy reports from parents sometimes differ from patient self-reports, especially when it comes to internalizing symptoms such as pain, fatigue, and emotional distress.^77,78 Parental anxiety can also affect their report of their child’s pain.⁷⁹

Some of the literature of pain evaluation focuses on pain intensity scales, which have been necessary for quantitating pain for the purposes of research and quality improvement efforts. In reality, however, each clinician faced with a patient experiencing pain integrates multiple data points from various sources in evaluating each patient. As an example, although self-report pain scores have classically been the gold standard in pain intensity assessment, recent observations and criticisms have led to a reevaluation of this dogma.⁸⁰ An isolated self-report pain score can have wide variability in meaning from patient to patient and can even be biased depending on situation, age, cognitive development, or experience. This leads to difficulty in translating overly simplistic pain scores directly into clinical decisions.

Instead, although pain is inherently a subjective experience and thus self-report is an important piece of information, bundled⁸¹ or hierarchical⁸² pain assessment approaches hint at ways to include other factors to inform clinical judgment. A bundled approach takes self-report, observation, proxy report, and other factors into account before making a global assessment of pain. One bundled approach, “CARES,” uses the following factors: Context (evaluating for likely sources of pain), Assess pain expression (including self-report and observation of pain behaviors and functional limitations from pain), Risk (balancing adverse effects of treatment with the clinical situation of the patient), Emotional factors (considering developmental and psychological factors), and Sociocultural factors (understanding patient and family preferences).⁸¹

Hierarchical assessments take the bundled approach a step further, by arranging pain-modifying factors into a specified order. An example of this is a pain assessment guideline by Herr et al.,⁸² developed for patients who cannot self-report pain reliably, including several groups which are often encountered in pediatric oncology—preverbal infants and toddlers, critically ill or unconscious patients, patients with intellectual disabilities, and some patients at the end of life. This particular hierarchical approach takes the following steps in order: (1) (attempt to) obtain self-report, (2) search for potential causes of pain, (3) observe patient behavior, (4) obtain proxy report, and (5) attempt an analgesic trial if pain behaviors continue despite providing for basic needs and comfort.

After consideration of all these factors, and a discussion with the child, caretakers, bedside providers, as well as the multidisciplinary team, a targeted treatment plan can then be developed. Reassessment after any analgesic trial is mandatory to further adjust the treatment plan to the patients’ needs.

Finally, it is worth mentioning that there has been work done to develop technologies to address barriers of inadequate assessment and documentation of pain and patient reluctance to report pain. Phone text messaging has been used to improve patient compliance with reporting pain intensity, duration, and pain-related disability.⁸³ A smartphone application similarly has been shown to be feasible and perform consistently compared to traditional interviewer-administered questionnaires.^81,82 The NIH PROMIS can be used on multiple platforms. Studies have shown its feasibility and cross-cultural validity for measuring functional mobility, pain interference, fatigue, depression, anxiety, peer relationships, and anger.^84,85 Tablet applications not only improve patient reporting of pain and symptoms but can also incorporate basic cognitive and behavioral skills training through electronic games.⁸⁶

A child undergoing cancer treatment will undergo many invasive, painful procedures. These include diagnostic or therapeutic procedures such as bone marrow biopsies, lumbar punctures (LPs), and surgery for tissue biopsy or tumor removal; venipunctures and subcutaneous port access (puncture) for diagnostic testing and/or treatment administration; placement of feeding tubes and urinary catheters; and placement of tunneled, peripherally inserted, or subcutaneous implanted “port” central venous access catheters. Even dressing changes (for central lines, surgical wounds) and suture and staple removal are a source of pain and distress for patients.¹⁰⁴

Not only does a child newly diagnosed with cancer face many repeated painful stimuli but these pain-evoking procedures can also occur in a relatively short period of time (within a few days, weeks, or months). In addition to the physical pain that invasive procedures cause related to tissue damage by insertion of a needle or device through the skin and/or bone, the procedures themselves produce a great deal of psychological distress, including fear and anxiety.¹⁰⁵ This anxiety can result in a quick recall of the procedure that will impact future pain management for all future painful procedures. Children with ALL on current treatment protocols, for example, undergo LPs on a regular basis ranging from once or twice weekly during induction therapy to once every 3 months during maintenance therapy. The initial LPs that are done at time of diagnosis and for the initial induction treatment therefore set the stage for the pain anticipated and experienced for children during their 2 to 3 years of ongoing chemotherapy. Studies in leukemia patients have clearly demonstrated that children have accurate memories of their painful procedures. The more negative a memory a child had about a previous LP, the higher the likelihood of increasing distress related to future LPs.¹⁰⁶ Similar findings have been found for children undergoing other repeated painful procedures.^107,108

There may also be specific groups of pediatric cancer patients that are at higher risk for distress due to painful, invasive procedures. Early studies have suggested that differences in reactions to painful stimuli can be attributed, at least in part, to a child’s temperament, including the dimensions of distractibility and persistence.^{109,110,111,112} Having a higher level of pain sensitivity (i.e., pain perception) is associated with greater anxiety and pain both prior to and during an LP procedure.¹¹³ In addition, the psychological stress and corresponding coping experienced by the parent can affect the child’s coping responses to the painful stressor. It has been shown that mothers of childhood cancer survivors do experience posttraumatic stress symptoms well into the survivorship period years after their child was treated for cancer.¹¹⁴ Thus, given that a child is dependent on parents for both physical and emotional support throughout the cancer care continuum, a child can be at increased risk for distress due to pain if his or her caretaker is not able to soothe or provide a safe, consistent environment because of his or her own stress and maladaptive coping.

Topical and local therapies for pain control are important. More than 50% of children report pain during venipunctures or intravenous cannulation,¹¹⁵ making these procedures an important target for pain prevention and control. Topical anesthetics are attractive because of the effectiveness without need for intravenous access and relative lack of systemic side effects. Eutectic mixture of lidocaine and prilocaine (eutectic mixture of local anesthetics [EMLA]) has been standard practice for decreasing procedural pain in children, available both in cream¹¹⁶ and patch¹¹⁷ forms. A warm lidocaine and tetracaine patch has also been shown to be effective for facilitating first-time needle procedure success.¹¹⁸ If insufficient time is available to allow topical anesthetic creams to take effect, vapocoolant sprays may be helpful in decreasing pain and decreasing intravenous cannulation failures in children.¹¹⁹ If anxiety is a major factor, low-dose oral midazolam has been useful in reducing fear and distress in younger children undergoing needle procedures such as subcutaneous port access.¹²⁰ High-dose acetaminophen¹²¹ or morphine¹²² did not have the same effect.

EMLA may also be useful by itself for LPs but only with those patients who undergo successful LP on their first attempt.¹²³ When used along with sedation, EMLA is useful in decreasing propofol use during LP.¹²⁴ When sedation is contraindicated, such as for patients with a large mediastinal mass with risk for airway compromise, topical and local anesthetic may be the only safe pain prevention strategy for these patients undergoing LP or bone marrow biopsy.

More and more pediatric oncology centers have instituted sedation and analgesia protocols for pediatric oncology procedures such as LPs and bone marrow biopsies and aspirations. Guidelines for pediatric procedural sedation have been published.^125,126 The desired level of sedation (e.g., anxiolysis vs. deep sedation vs. general anesthesia) depends on patient age and cognitive development and painfulness of the procedure, balanced with the risks of sedation (e.g., in setting of airway compression from a mediastinal mass).¹²⁷ Level of sedation also depends on the need for immobility, as there is a risk of introducing leukemic cells from blood into the spinal fluid with traumatic LP during induction therapy for ALL.¹²⁸