The aim of this study was to identify the best sedation/analgesia protocol for laser photocoagulation (PC) of retinopathy of prematurity (ROP).
This multicenter observational study included five hospitals, each using a specific sedation/analgesia protocol: local anesthesia with oxybuprocaine hydrochloride (Group L); intravenous pentazocine (Group P); intravenous fentanyl (Group F); air, oxygen and sevoflurane (AOS) inhalation (Group I). The groups were compared for pain responses, vital signs and adverse events.
Heart rates and systemic blood pressures were elevated by PC in Groups L and P and Groups L, P and F, respectively. Moreover, poor analgesic efficacy was recognized in Groups L, P and F. In contrast, Group I experienced hypothermia, enteral feeding intolerance and apnea more frequently.
From the viewpoint of sedation/pain relief, AOS anesthesia should be the best protocol. However, considering all the various factors together, the most reasonable one can be varied based on the patient’s condition and hospital.
Retinopathy of prematurity (ROP) is a disease of premature babies, characterized by abnormal blood vessel development in the retina. Some cases resolve spontaneously, whereas others may lead to proliferation, retinal detachment or even blindness. As the number of very low birth weight infants, which is <1500 g, and their survival rate are increasing, we have more opportunities to perform laser photocoagulation (PC) for ROP, which is the most common procedure for treatment. ROP-PC may cause significant systemic stress and potentially life-threatening cardiorespiratory events.1 There is a variation in the protocols for sedation/analgesia for PC of ROP in neonatal care units; however, there is no consensus or guideline on this issue. For example, a survey conducted in the United Kingdom showed considerable variation: 50% of ophthalmologists use general anesthesia, whereas 37% use intravenous sedation combined with topical anesthesia. Other methods include oral sedation with topical anesthesia and rectal chloral hydrate and paracetamol combined with topical anesthesia.2
The central nervous system of infants is sufficiently developed to perceive pain anatomically, neurophysiologically and neurochemically.3, 4 Inadequate sedation/analgesia can lead to neurodevelopmental late effects as well as ethical problems. The experience of pain during the neonatal period has been reported to increase the intensity of reaction to pain in later infancy.5, 6 Also, there are many reports on the effects of sedative/anesthetic drugs on the central nervous system, including inhalation anesthetics, raising serious concerns for neurodevelopment, especially in neonatal medicine.7, 8, 9, 10 Therefore, a most suitable protocol for sedation/analgesia for PC of ROP should be one with less pain, which may have a minimal effect on neurodevelopment in neonates.
This multicenter observational study compared the four most commonly used sedation/analgesia protocols in Japan for PC of ROP in terms of pain reduction as a primary outcome, vital signs (such as heart rate and blood pressure), operation duration and adverse events during and after the procedure as secondary outcomes.
This multicenter observational study was conducted in five hospitals: Nagoya University Hospital, Japanese Red Cross Nagoya Daiichi Hospital, Anjo Kosei Hospital, Ogaki Municipal Hospital and Toyota Memorial Hospital.
Parental informed consents and approval by the local ethics committee in each hospital were obtained, and this study followed the tenets of the Declaration of Helsinki. This study included only those patients who required laser therapy for ROP according to the following criteria: (i) zone І, any stage ROP with plus disease, (ii) zone І, stage 3 ROP without plus disease and (iii) zone II, stage 2 or stage 3 ROP with plus disease, according to the international classification.11 The patients whose parents refused to give informed consents were excluded from this study. The laser therapies were performed by an ophthalmologist in all hospitals under the same treatment protocol. In case of patients who required more than one course of laser therapy, we evaluated only the first therapy. This study was performed prospectively from December 2009 to December 2010. In addition, owing to the lack of number of patients, the data in the patients receiving air, oxygen and sevoflurane (AOS) inhalation between October 2007 and November 2009 were collected retrospectively as well.
Each hospital selected one of the following sedation/analgesia protocols: only local anesthesia with oxybuprocaine hydrochloride, without any other intravenous or inhalation anesthesia (Group L); intravenous pentazocine and midazolam (Group P); intravenous fentanyl and midazolam (Group F); or AOS inhalation (Group I). Oxybuprocaine hydrochloride was used as a local anesthetic by an ophthalmologist in all hospitals.
In Groups P and F, intravenous sedation/analgesia was performed by a neonatologist in the neonatal intensive care unit. In both groups, atropine sulfate was used as needed. The dosage of each analgesic was as follows: pentazocine, 0.5 mg kg−1; fentanyl, 5 μg kg−1; midazolam, 0.1 mg kg−1; and atropine sulfate, 0.01 mg kg−1. When these doses were not sufficient to control pain during laser therapy, an additional half volume of pentazocine or fentanyl and/or one-fourth volume of midazolam were administered. The number of additional administrations was unlimited. For Group I, AOS anesthesia was performed with 2 to 5% sevoflurane by an anesthetist in the operation room.
Evaluation of pain responses and vital signs
Pain relief during PC was evaluated as a primary outcome with the Neonatal Infant Pain Scale (NIPS) (Supplementary Table 1).12 This tool measures behavioral signs of pain, including facial expression, cry, breathing patterns, movement of arms and legs and state of arousal, in addition to heart rate and oxygen saturation changes.
The following vital signs were monitored before, during and after PC: heart rate, blood pressure, oxygen saturation (SpO2), body temperature and urine volume. The groups were also compared for operation duration, and any adverse events during and after PC.
Heart rate, blood pressure and SpO2 were recorded every 10 min during PC, and their mean values were calculated. In addition, the mean values for these parameters were calculated 8 h before and after PC; the values were recorded every 1 to 2 h. The adverse events evaluated in the present study included oxygen desaturation (<70%) and bradycardia (<100/min) during PC; hypothermia (<36.5 °C) during and after PC; and enteral feeding intolerance (24 h), oliguria (over 50% reduction) and apnea (>150% frequency of pre-PC/>6 times a day in infants who had no apnea before PC) after PC.
All analyses were performed using IBM SPSS Statistics 20 (SPSS IBM Japan, Tokyo, Japan), and all data are expressed as the mean±s.e.m. Statistical analysis was performed by one-way analysis of variance followed by Scheffe’s post hoc test. χ2 test with Bonferroni correction was used to compare prevalence. A P-value of <0.05 were considered significant.
Demographics of participants
One hospital used only local anesthesia (Group L), three used intravenous medication (Group P or F) and one used AOS inhalation (Group I). The number of patients enrolled in the present study was 15, 11, 11 and 12 in Groups L, P, F and I, respectively. Table 1 compares the demographics of the four patient groups. There was no significant difference in birth body weight, although the gestational age (weeks per days) of Group I (24/0) was significantly lower compared with that of Group F (26/4) (P<0.05). The prevalence of chronic lung disease, defined as requirement for supplemental O2 at 36 weeks postmenstrual age, was 36% in Group P, which was significantly lower compared with that in Group L (87%, P<0.05) or Group I (100%, P<0.01). Postmenstrual ages (weeks per days) at PC were significantly longer in Group P (35/2) and Group F (37/2) compared with that in Group L (33/4). Body weights at PC were significantly lower in Group L (1218 g) compared with that in Group P (1772 g). All patients in Group I were on a respirator at the time of PC compared with 53%, 36% and 91% in Groups L, P and F, respectively (Group L vs Group I, P<0.05; Group P vs Group I, P<0.01). In contrast, all groups exhibited comparable prevalence of intraventricular hemorrhage (⩾grade 2; defined as any hemorrhage extending beyond the germinal matrix), hemoglobin and FiO2.
The patient groups were also compared in terms of pain response to PC and adverse events. The NIPS scores revealed strong pain responses to PC in Group L (mean, 7.0±1.9; max, 8.3±1.3), which were higher compared with those in Groups P and F (Figure 1; all P<0.01, both mean and max). In contrast, no pain response was detected in Group I (Figure 1; P<0.01, Group L vs Group I and P<0.05, Group P vs Group I).
Sedative and analgesic effects on vital signs
The impact of each sedation/anesthesia protocol on vital signs was determined in terms of heart rate and systemic blood pressure measured before and during PC. Heart rate measured during PC was significantly higher in Group L compared with that in Groups F and I (Figure 2a; all P<0.01). After PC, Group L had significantly higher heart rates compared with the other groups (Figure 2a; P<0.01, Group L vs Group F and Group I; P<0.05, Group L vs Group P). The heart rates of Groups F and I were not affected by PC, whereas those of Groups L and P were elevated by 16.6% and 9.8%, respectively, compared with pre-PC values (Figure 2b; P<0.01, Group L vs Group F and Group I). Systemic blood pressure during PC was significantly lower in Group I compared with that in the other groups. The procedure raised blood pressures in Groups L, P, and F by >20%, whereas the values decreased by 10% in Group I (Figure 3a and b; all P<0.01).
The adverse events recorded during and after PC are listed in Table 2. Desaturation and bradycardia were seen more often in Groups P and F (desaturation, P<0.01, Groups P and F vs Group I; bradycardia, P<0.05, Group P vs Group L and Group F vs Group I). In contrast, hypothermia during PC developed more often in Group I (P<0.01 and 0.05, Group I vs Group L and P, respectively; P<0.05, Group F vs Group L). Apnea and enteral feeding intolerance after PC tended to develop more often in Group I, but did not reach significant difference. Oliguria developed in Groups P, F and I, with significant differences between Groups F and L (P<0.01).
Operation durations and repetitions
The PC procedure tended to be longer in Group I, but no significant difference was found between the groups (Figure 4). Additional PC sessions were required in 73% of Group L, 45% of Group P and 55% of Group F, but not in Group I (P<0.01, Group I vs Groups P and L; P<0.05, Group I vs Group F).
The PC treatment to the avascular area in ROP (surgical repair of abnormal retinal blood vessels in premature neonates by PC) is an extremely delicate procedure that relies on the immobility of the patient. As such, the identification of an efficient sedation/analgesia protocol for pain relief is critical to ensure a positive treatment outcome. The present study provides the first comparative analysis of the commonly used sedation/analgesia protocols for the treatment of ROP by PC.
Inhalation AOS anesthesia (Group I) was the only sedation/analgesia protocol that did not affect the vital signs during PC, and provided adequate pain relief. The lack of body movement might facilitate the procedure and allow longer operation time; thus, no additional PC was required to complete the treatment. On the other hand, gas inhalation requires scheduling an operating room and an anesthetist within the narrow window of opportunity for effective treatment after birth. Only few hospitals in Japan have a sufficient number of anesthetists. Furthermore, moving a small and/or unstable patient to the operating room may lead to hypothermia and/or some other condition that is more unstable. We demonstrated that AOS anesthesia was associated with a higher prevalence of hypothermia compared with the other protocols. In addition to hypothermia, enteral feeding intolerance or oliguria developed more often in AOS anesthesia. Our findings suggest that adverse events after PC are more likely to develop in Group I. Furthermore, it is very important to understand that gas inhalation anesthesia may have negative effects on the brain.8, 9, 10
Intravenous sedation/analgesia can be performed in the NICU without having to move the patient. Because there is no need to schedule an operating room, PC can be initiated sooner, with easier follow-up procedures. Furthermore, the fact that intubation is not necessary is a clear asset because most premature babies, which requires PC, are with chronic lung disease. Intubation itself can sometimes be an invasive procedure and can lead to difficulty in extubation after PC, particularly in patients with chronic lung disease.13, 14 In addition, intravenous sedation/analgesia in the NICU can be performed by anesthetists or neonatologists.13, 15 However, based on NIPS evaluation, our study showed that intravenous sedation/analgesia does not provide sufficient pain relief. The NIPS scores were particularly high with pentazocine. Furthermore, these patients experienced oxygen desaturation and bradycardia during PC more often compared with the other groups. Moreover, we cannot ignore the possible negative effect of intravenous sedation/analgesia on the developing brain.7, 8, 9
Local sedation/analgesia without any general sedation/analgesia avoids the possible adverse effects of inhaled/intravenous agents, particularly the effects on the brain. Furthermore, PC can be performed in the NICU without anesthetists or intubation. We reported fewer postprocedure adverse events compared with the other sedation/analgesia protocols. However, local analgesia alone did not provide adequate pain relief. Haigh et al.1 reported that PC with only local anesthesia may induce life-threatening severe respiratory and/or circulatory failure. In addition, the experience of intense pain in the neonatal period may lead to neurodevelopmental late effects.6 In 2006, the American Academy of Pediatrics and Canadian Pediatric Society advocated that sedation/analgesia for PC with only oral sucrose or local anesthetics is not adequate,16 although oral sucrose and non-nutritive sucking goes some way to reducing pain.17 In United Kingdom, PC was performed only with local anesthetics in 23% of the hospitals in 1993,18 although none used this protocol in 2003.2
The present study has some limitations. Because the patients were not randomized, the four groups differed with respect to certain parameters: gestational age, prevalence of chronic lung disease, postmenstrual ages and body weight at PC. Therefore, we reanalyzed after adjusting for these parameters using multiple regression analysis. However, there was still a significant difference in scores using NIPS and changes in the heart rate/blood pressure (data not shown). This is a multicenter study. Thus, several factors may contribute to intergroup variability, apart from the sedation/anesthesia protocol, including different NICU personnel, ophthalmologists and anesthetists.
The present study compared the common sedation/anesthesia protocols currently used for PC procedures in neonates. From the viewpoint of sedation/pain relief, inhalation AOS anesthesia should be the best protocol for PC of ROP. However, considering all the various factors together, the most reasonable one can be varied based on the patient’s condition and hospital. The method for sedation/analgesia should be selected on the basis of the health condition of each patient (i.e., chronic lung disease), the inherent risks associated with moving the patient from the NICU to the operating room and the personnel limitations of the establishment.
Haigh PM, Chiswick ML, O'Donoghue EP . Retinopathy of prematurity: systemic complications associated with different anaesthetic techniques at treatment. Br J Ophthalmol 1997; 81 (4): 283–287.
Chen SD, Sundaram V, Wilkinson A, Patel CK . Variation in anaesthesia for the laser treatment of retinopathy of prematurity—a survey of ophthalmologists in the UK. Eye (Lond) 2007; 21 (8): 1033–1036.
Anand KJ, Carr DB . The neuroanatomy, neurophysiology, and neurochemistry of pain, stress, and analgesia in newborns and children. Pediatr Clin N Am 1989; 36 (4): 795–822.
Giannakoulopoulos X, Sepulveda W, Kourtis P, Glover V, Fisk NM . Fetal plasma cortisol and beta-endorphin response to intrauterine needling. Lancet 1994; 344 (8915): 77–81.
Taddio A, Goldbach M, Ipp M, Stevens B, Koren G . Effect of neonatal circumcision on pain responses during vaccination in boys. Lancet 1995; 345 (8945): 291–292.
Taddio A, Katz J, Ilersich AL, Koren G . Effect of neonatal circumcision on pain response during subsequent routine vaccination. Lancet 1997; 349 (9052): 599–603.
Laudenbach V, Calo G, Guerrini R, Lamboley G, Benoist JF, Evrard P et al. Nociceptin/orphanin FQ exacerbates excitotoxic white-matter lesions in the murine neonatal brain. J Clin Invest 2001; 107 (4): 457–466.
Loepke AW, Soriano SG . An assessment of the effects of general anesthetics on developing brain structure and neurocognitive function. Anesth Analg 2008; 106 (6): 1681–1707.
Mellon RD, Simone AF, Rappaport BA . Use of anesthetic agents in neonates and young children. Anesth Analg 2007; 104 (3): 509–520.
Zhu C, Gao J, Karlsson N, Li Q, Zhang Y, Huang Z et al. Isoflurane anesthesia induced persistent, progressive memory impairment, caused a loss of neural stem cells, and reduced neurogenesis in young, but not adult, rodents. J Cereb Blood Flow Metab 2010; 30 (5): 1017–1030.
International Committee for the Classification of Retinopathy of Prematurity. The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol 2005; 123 (7): 991–999.
Lawrence J, Alcock D, McGrath P, Kay J, MacMurray SB, Dulberg C . The development of a tool to assess neonatal pain. Neonatal Netw 1993; 12 (6): 59–66.
Parulekar MV, Chen SD, Patel CK . Sub-tenon's local anaesthesia for the treatment of retinopathy of prematurity with diode laser. Eye (Lond) 2008; 22 (3): 375–379.
Woodhead DD, Lambert DK, Molloy DA, Schmutz N, Righter E, Baer VL et al. Avoiding endotracheal intubation of neonates undergoing laser surgery for retinopathy of prematurity. J Perinatol 2007; 27 (4): 209–213.
Kirwan C, O'Keefe M, Prendergast M, Twomey A, Murphy J . Morphine analgesia as an alternative to general anaesthesia during laser treatment of retinopathy of prematurity. Acta Ophthalmol Scand 2007; 85 (6): 644–647.
Batton DG, Barrington KJ, Wallman C . Prevention and management of pain in the neonate: an update. Pediatrics 2006; 118 (5): 2231–2241.
Dilli D, Ilarslan NE, Kabatas EU, Zenciroglu A, Simsek Y, Okumus N . Oral sucrose and non-nutritive sucking goes some way to reducing pain during retinopathy of prematurity eye examinations. Acta Paediatr 2014; 103 (2): e76–e79.
Schulenburg WE, Bloom PA . Current problems in the management of ROP. Acta Ophthalmol Scand Suppl 1995; 214: 14–16.
We are grateful to Dr Tetsuo Hattori, Dr Reina Hyodo and Dr Yoshihiro Tanahashi for collecting the data. The authors have not received any financial support for the present study.
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on the Journal of Perinatology website
About this article
Cite this article
Sato, Y., Oshiro, M., Takemoto, K. et al. Multicenter observational study comparing sedation/analgesia protocols for laser photocoagulation treatment of retinopathy of prematurity. J Perinatol 35, 965–969 (2015). https://doi.org/10.1038/jp.2015.112
Low dose fentanyl infusion versus 24% oral sucrose for pain management during laser treatment for retinopathy of prematurity—an open label randomized clinical trial
European Journal of Pediatrics (2019)
Literature review informs clinical guidelines for pain management during screening and laser photocoagulation for retinopathy of prematurity
Acta Paediatrica (2019)
Survey of Ophthalmology (2017)