Free
Clinical and Epidemiologic Research  |   May 2014
Natural Involution of Acute Retinopathy of Prematurity Not Requiring Treatment: Factors Associated With the Time Course of Involution
Author Notes
  • Department of Ophthalmology, Eye & ENT Hospital of Fudan University, Shanghai, People's Republic of China 
  • Correspondence: Xin Huang, Department of Ophthalmology, Eye & ENT Hospital of Fudan University, Shanghai 200031, People's Republic of China; xinhuang66@126.com
Investigative Ophthalmology & Visual Science May 2014, Vol.55, 3165-3170. doi:10.1167/iovs.13-13744
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Ying-Qin Ni, Xin Huang, Kang Xue, Jia Yu, Lu Ruan, Hai-Dong Shan, Ge-Zhi Xu; Natural Involution of Acute Retinopathy of Prematurity Not Requiring Treatment: Factors Associated With the Time Course of Involution. Invest. Ophthalmol. Vis. Sci. 2014;55(5):3165-3170. doi: 10.1167/iovs.13-13744.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: We identified the timing of natural involution of acute retinopathy of prematurity (ROP) not requiring treatment and determined the risk factors associated with delayed involution.

Methods.: In this retrospective case series, 82 eyes (the more severe eye) of 82 infants who developed at least one clock hour of acute ROP, stages 1 through 3, but who didn't progress to type 1 ROP, were selected for analysis. The location, extent, and severity of ROP were documented by investigators during serial retinal examinations. The onset and completion of the ROP's involution were determined from a review of these data. Two groups were classified by the involution pattern: Group 1 included infants whose ROP disease involuted before 50 weeks of postmenstrual age, and Group 2 included infants whose ROP disease involuted over 50 weeks (delayed involution). A total of 14 possible risk factors was included in the logistic regression analysis to assess the relationship between each factor and the involution pattern.

Results.: Acute ROP not requiring treatment began to involute at a mean of 40.4 weeks of postmenstrual age and finished at a mean of 50.6 weeks. Involution began at the same mean postmenstrual age for each zone of disease (P = 0.48) and finished earlier in zone III than in zone II (P < 0.01). An analysis by severity of ROP found that involution began the earliest with the mildest disease (stage 1; mean, 38.1 weeks) and latest with the most serious disease (stage 3; mean, 42.3 weeks; P < 0.01). Zone II disease took longer to finish involution (16.04 ± 12.35 weeks) than zone III (8.30 ± 7.3 weeks), and stage 3 (23.88 ± 10.58 weeks) took longer to finish involution than stage 1 (2.03 ± 0.96 weeks) and stage 2 disease (7.69 ± 4.75 weeks, P < 0.01, respectively). No unfavorable outcome was found in our series. Multivariable logistic regression analysis showed that continuous positive airway pressure (CPAP, P < 0.0001), active stage 3 disease (P = 0.006), and anemia (P = 0.03) were significant risk factors associated with delayed involution.

Conclusions.: The natural involution of acute ROP not requiring treatment correlated better with severity than with ROP location. Active stage 3 disease, CPAP, and anemia were predictive risk factors for delayed involution of ROP.

Introduction
Retinopathy of prematurity (ROP) is a leading cause of childhood preventable blindness, particularly in the middle-income developing countries of Asia. 13 In China, the characteristics of infants with ROP are quite different from those in industrialized countries, because bigger and more mature infants also develop severe ROP. 4,5 The Chinese Ministry of Health developed guidelines for the prevention and treatment of ROP, published in 2004, to be implemented by pediatricians and ophthalmologists. 6 Increased awareness of the risk of blindness caused by ROP has led to the expansion of a comprehensive screening program, resulting in a significant decrease in the numbers of stage 4 and stage 5 ROP in China. 4  
Involution of ROP typically is characterized by a downgrading of staging and/or growth of retinal vessels into a more peripheral zone. 7 Although a small number of active involuted ROP cases progress to unfavorable retinal outcomes, most eyes reach relatively stable states known as regressed ROP. 8 Repka et al. 7 reported that 90% of acute-phase ROP cases began to involute before 44 weeks postmenstrual age; however, little is known about the time of the ultimate resolution of involution. On the other hand, although extensive studies have been conducted regarding the risk factors of ROP, 912 no formal studies on the risk factors of delayed involution have been conducted to our knowledge. Understanding the regular pattern of involution and possible predictive factors associated with delayed involution is quite important to ophthalmologists. Such information might be vital to determine the most efficient and cost-effective management strategy for ROP screening in China. It also might provide guidance for ophthalmologists to schedule individualized serial examination for infants who should be followed up for a longer time. 
The purpose of our study was to determine the regular pattern of involution of acute ROP not requiring treatment, including the timing of the onset, completion, and duration of involution. We also examined a series of suspected postnatal influences to identify independent risk factors associated with the delayed involution of ROP. 
Patients and Methods
We enrolled in this study 82 infants admitted to the Eye & ENT Hospital of Fudan University between March 2011 and April 2012, with birth weights (BWs) ≤ 2000 g or gestational age (GA) ≤ 34 weeks. The infants were examined according to the 2004 Chinese Ministry of Health guidelines, 6 and a wide range of criteria was used to avoid missing any infants with treatable ROP disease. The first eye examination was performed at the postnatal age of 4 weeks. The research followed the tenets of the Declaration of Helsinki and written informed consent was obtained from each patient's parents for the sequential examinations that provided data for this study. On each examination, dilation was performed using phenylephrine 2.5% and cyclopentolate 0.5%. A RetCam (Clarity Medical Systems, Pleasanton, CA, USA) wide-angle fundus imaging system was used to record an image of the ocular fundus. The examinations were performed on a weekly, biweekly, or monthly schedule, depending on the ocular findings, and continued until it was established that the ROP was definitely finishing involution. The ROP was classified according to the revised International Classification of Retinopathy of Prematurity, including the stage, extent, zone, and presence or absence of “plus” or “preplus” disease. 13  
All of the 82 infants developed at least one clock hour of acute ROP, stages 1 through 3, but it did not progress to type 1 ROP. Only the eye with the more advanced stage of ROP from each patient was designated for analysis, because the data from the two eyes could not be considered to be independent. Meeting any one of the following criteria was sufficient to define the onset of involution, in accordance with the study of Repka et al. 7 : (1) a reduction by two or more sectors in the number of sectors of the highest stage of acute disease recorded in a given zone from one examination to the next, and (2) documented completion of vascularization in all sectors of a zone that previously had acute ROP. Meeting any one of the following criteria was sufficient to define the clinical completion of involution: (1) complete regression of acute ROP with vascularization of zone III and/or (2) regressed ROP with peripheral retinal changes of pigmentary changes, vitreoretinal interface changes, and vitreous membrane, with or without attachment to retina, but with the retinal vessels stepping over the region that previously had acute ROP and growth into a more peripheral zone. Serial research examinations were performed approximately every two weeks (one week for type 2 ROP) before 45 weeks postmenstrual age and monthly after 45 weeks postmenstrual age until the defined criteria of the clinical completion of involution were met. The age of onset and completion of involution was defined as the mean age between the two serial retinal examinations that showed acute ROP at the earlier examination and definite onset or completion of involution. 
A total of 14 potential risk factors, including ocular manifestation of the stage, zone, and “preplus” of ROP, BW, GA, and asphyxia at birth, neonatal respiratory distress syndrome, sepsis (positive blood culture), hyperbilirubinemia (defined as total serum bilirubin concentration > 15 mg/dL), anemia (defined as Hb < 110 g/L),and acidosis occurring during the first four weeks after birth, was analyzed with logistic regression aimed at adjusting interaction between the variables and possible confounding factors for identifying independent risk factors for the delayed involution of ROP. The criterion for dropping variables from the model during backward stepwise logistic regression (Wald) was a P value ≥ 0.05. 
Student's t-test was used to investigate the influences of different zones of ROP on the onset, completion, and duration times of involution. The 1-way ANOVA was used to analyze the influences of different stages, BWs, and GAs on the onset, completion, and duration times of involution. Statistical analysis was performed using SPSS 13.0 for Windows (SPSS, Inc., Chicago, IL, USA). A P value of <0.05 (2-sided) was considered significant for all tests. 
Results
This study included 82 consecutive infants whose selected eye developed acute ROP not requiring treatment and who completed our follow-up assessment. Of the infants 55 (67%) were males. The mean estimated GA was 30.5 weeks (range, 26.2–35.5), and the mean BW was 1491.6 g (range, 840–2350). Of the 82 eyes 61 (74.4%) had zone III disease and 21 (25.6%) had zone II disease; 17 (20.7%) of the 82 eyes had stage 3 disease, 54 (65.9%) had stage 2 disease, and 11 (13.4%) had stage 1 disease. 
Involution generally began between 36.5 and 44.4 weeks (5%–95%) postmenstrual age. The mean time of onset was 40.4 weeks postmenstrual age. The time of onset of involution based on postmenstrual age was analyzed in terms of zone and stage of ROP. The zone of ROP did not affect the timing of onset of involution, because involution began at virtually the same mean postmenstrual age for each zone of disease (Table 1, P = 0.48). The analysis by stage of ROP found clinically important differences in onset times of involution (Table 2). Involution was found to begin earliest with the mildest disease (stage 1; mean, 38.1 weeks) and latest with the most severe disease (stage 3; mean, 42.3 weeks); this finding was statistically significant (P < 0.01). 
Table 1
 
Start Point of Involution by Zone of Acute ROP
Table 1
 
Start Point of Involution by Zone of Acute ROP
Zone No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
II 21 40.6 (37.1–45.1)
III 61 40.3 (36.3–44.4)
Total 82 40.4 (36.5–44.4)
Table 2
 
Start Point of Involution by Stage of ROP
Table 2
 
Start Point of Involution by Stage of ROP
Stage No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
1 11 38.1 (34.5–39.6)
2 54 40.3 (37.2–44.6)
3 17 42.3 (39.6–45.2)
Total 82 40.4 (36.5–44.4)
According to our criteria, the time of the completion of involution generally was between 39.0 and 75.0 weeks (5%–95%) postmenstrual age, with a mean time of 50.6 weeks. Zone III disease finished involution at a mean postmenstrual age of 48.5 weeks, which was much earlier than zone II disease, with a mean age of 56.6 weeks (Table 3, P < 0.01). The analysis of the duration time of involution showed that zone II disease took longer to finish involution (16.04 ± 12.35 weeks) than zone III disease (8.30 ± 7.37 weeks). By stage of ROP, our results demonstrated that involution finished latest with the most severe disease (stage 3; mean, 66.0 weeks) and earliest with the mildest disease (stage 1, 39.6 weeks, Table 4). Stage 3 disease (23.88 ± 10.58 weeks) took longer to finish involution than stage 1 (2.03 ± 0.96 weeks) and stage 2 disease (7.69 ± 4.75 weeks, P < 0.01, respectively). 
Table 3
 
End Point of Involution by Zone of Acute ROP
Table 3
 
End Point of Involution by Zone of Acute ROP
Zone No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
II 21 56.6 (39.2–84.2)
III 61 48.5 (37.5–70.4)
Total 82 50.6 (39.0–75.0)
Table 4
 
End Point of Involution by Stage of Acute ROP
Table 4
 
End Point of Involution by Stage of Acute ROP
Stage No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
1 11 39.6 (36.5–42.2)
2 54 48.3 (40.6–59.5)
3 17 66.0 (48.4–84.1)
Total 79 50.6 (39.0–75.0)
The analysis by BW showed that the onset of involution was within the same postmenstrual week. The BW < 1000 g group took a longer time to finish involution than did the other groups (P < 0.01). Analysis by GA revealed a longer duration time for involution in the GA < 28 weeks group (P = 0.046). 
In our series, all of the 82 eyes with acute ROP not requiring treatment underwent spontaneous involution without unfavorable outcomes. Generally speaking, the involution of acute ROP not requiring treatment was a self-limiting process characterized by the downgrading of staging and/or growth of retinal vessels into a more peripheral zone. However, the clinical features of involution were quite different in ROP disease with different stages. The involution of stage 1 ROP was characterized by the gradually blurring and disappearing demarcation line (Figs. A–C). For stage 2 ROP, the elevated ridge flattened and the retinal vessels grew into a more peripheral zone (Figs. D–I). In stage 3, neovascularization or extraretinal fibrovascular proliferation grows from the ridge. During the involution, the neovascularization regressed with a flattening ridge and the retinal vessels stepped over the region that previously had acute ROP and grew into a more peripheral zone. However, involutional sequelae, such as vitreoretinal interface changes, may be seen in the area of the previously active stage 3 disease. (Figs. J–O
Univariable and multivariable logistic regression models were conducted to identify the factors associated with the delayed involution, defined as a completion time of involution of over 50 weeks postmenstrual age. Four factors, including GA < 28 weeks, asphyxia, hyperbilirubinemia, and BW < 1000 g, were not statistically associated with delayed involution in the univariable model (Table 5). Neonatal respiratory distress syndrome (NRDS), preplus disease, sepsis, oxygen, acidosis, zone II disease, and male sex were significantly associated with delayed involution in the univariable model, but not in the multivariable model (Table 5). Continuous positive airway pressure (CPAP), active stage 3 disease, and anemia were found to be statistically associated with delayed involution. In the multivariable logistic regression model, CPAP increased the odds of delayed involution by a factor of 20.50 (4.55∼101.60, P < 0.0001), active stage 3 disease increased the odds by a factor of 16.60 (2.27∼121.09, P = 0.006), and anemia increased the odds by a factor of 4.99 (1.12∼22.29, P = 0.03, Table 6). Our results indicated that these three factors were predictive risk factors for delayed involution of ROP. 
Figure
 
Fundus photograph to illustrate the natural involution of mild acute ROP. (AC) Involution of stage 1 disease. View of the demarcation line (white arrows) which gradually blurred and disappeared by 42 weeks postmenstrual age. (DI) Involution of stage 2 disease. The ridge (white arrows) at the junction between vascularized and avascular retina was the hallmark of stage 2 ROP. Small isolated tufts, commonly called “pop-corn” (black arrows), were seen posterior to the ridge. During the involution process, the elevated ridge flattened, the “pop-corn” disappeared, and the retinal vessels grew into a more peripheral zone. Vascularization of the retina was completed by 49 weeks postmenstrual age. (JO) Involution of mild stage 3 disease. A stage 2 disease (ridge) was found at 39 weeks postmenstrual age and progressed to stage 3 at 40 weeks with neovascularization extended from the ridge into the vitreous in the inferior temporal region (black arrow). Note the tortuosity and dilatation of posterior pole vessels that were insufficient for plus disease, which we called “preplus” disease (red arrows). The extraretinal fibrovascular tissue regressed spontaneously after 2 weeks with a flattening ridge. The retinal vessels stepped over the region that previously had acute ROP and grew into a more peripheral zone at 62 weeks postmenstrual age. However, vitreous condensation and vitreous membrane (yellow arrows) were seen in the area of previously active stage 3 disease.
Figure
 
Fundus photograph to illustrate the natural involution of mild acute ROP. (AC) Involution of stage 1 disease. View of the demarcation line (white arrows) which gradually blurred and disappeared by 42 weeks postmenstrual age. (DI) Involution of stage 2 disease. The ridge (white arrows) at the junction between vascularized and avascular retina was the hallmark of stage 2 ROP. Small isolated tufts, commonly called “pop-corn” (black arrows), were seen posterior to the ridge. During the involution process, the elevated ridge flattened, the “pop-corn” disappeared, and the retinal vessels grew into a more peripheral zone. Vascularization of the retina was completed by 49 weeks postmenstrual age. (JO) Involution of mild stage 3 disease. A stage 2 disease (ridge) was found at 39 weeks postmenstrual age and progressed to stage 3 at 40 weeks with neovascularization extended from the ridge into the vitreous in the inferior temporal region (black arrow). Note the tortuosity and dilatation of posterior pole vessels that were insufficient for plus disease, which we called “preplus” disease (red arrows). The extraretinal fibrovascular tissue regressed spontaneously after 2 weeks with a flattening ridge. The retinal vessels stepped over the region that previously had acute ROP and grew into a more peripheral zone at 62 weeks postmenstrual age. However, vitreous condensation and vitreous membrane (yellow arrows) were seen in the area of previously active stage 3 disease.
Table 5
 
Univariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Table 5
 
Univariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Risk Factor β OR 95% CI of OR P Value
CPAP 3.43 30.89 8.84∼107.95 <0.0001
NRDS 3.23 25.19 7.59∼83.62 <0.0001
Preplus disease 3.08 21.86 2.60∼183.48 0.004
Sepsis 2.74 15.52 1.80∼133.56 0.01
Active stage 3 disease < 5 clock-h 2.66 14.30 3.64∼56.17 <0.0001
Oxygen 2.48 12 2.59∼55.68 0.002
Anemia 2.29 9.87 3.36∼28.97 <0.0001
Acidosis 1.69 5.44 1.67∼17.78 0.005
Zone II disease 1.44 4.21 1.48∼11.94 0.006
Male 1.30 3.67 1.21∼11.09 0.02
GA < 28 wks 1.41 4.08 0.94∼17.75.20 0.06
Asphyxia 0.85 2.34 0.75∼7.28 0.14
Hyperbilirubinaemia 0.42 1.53 0.62∼3.77 0.36
BW < 1000 g 0.38 1.11 0.99∼1.25 0.08
Table 6
 
Multivariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Table 6
 
Multivariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Risk Factor β OR 95% CI of OR P Value
CPAP 3.07 20.50 4.55∼101.60 <0.0001
NRDS 1.04 2.84 0.22∼37.14 0.43
Preplus disease 1.15 3.17 0.10∼99.75 0.51
Sepsis −0.78 0.46 0.03∼7.89 0.59
Active stage 3 disease < 5 clock-h 2.81 16.60 2.27∼121.09 0.006
Oxygen 0.89 2.44 0.33∼18.07 0.38
Anemia 1.61 4.99 1.12∼22.29 0.03
Acidosis 0.97 2.64 0.44∼15.83 0.29
Zone II disease 1.45 4.28 0.76∼24.02 0.09
Male 0.84 2.32 0.39∼13.86 0.35
Discussion
The condition of ROP is known to undergo spontaneous involution in 90% of infants, which is considered a favorable clinical event. 7,8,14 The downgrading of staging and/or growth of retinal vessels into a more peripheral zone are the main characteristics of involution. 7 The process of involution occurs first at the junction of the vascular and avascular retinal regions when the disease moves from zone I to zone II or from zone II to zone III. 13,15 The color of the ridge might change from salmon pink to white, indicating a progressive tendency toward inactivity. 13,15 Understanding the detailed information about the timing of these events is a challenging but essential goal. 
The important research finding from the CRYO-ROP database by Repka et al. 7 focused on the onset time of involution. However, information about the completion of involution has not been available in the literature describing ROP, as it was not considered necessary that acute ROP infants be evaluated up to complete resolution. In our series, the 82 infants with acute ROP not requiring treatment underwent regular follow-up examinations until the defined criteria of the completion of involution were met. The reason for the lengthy and complete follow-up is that, as stated previously, the characteristics of ROP in China are different from those in developed countries, as bigger, more mature infants also develop severe ROP. 4,5 In our unpublished data, two infants with type 2 ROP who progressed to type 1 ROP after 45 weeks postmenstrual age and underwent laser treatment would have been left out according to the follow-up criteria recommended by Repka et al. 7 On the other hand, although acute ROP not requiring treatment usually involutes without progressing to unfavorable outcomes, residual cicatricial changes can remain. 1618 Eyes with lattice-like degeneration, vitreous condensation, and vitreoretinal interface changes are at high risk of developing retinal detachment; 19,20 therefore, ROP should be followed up throughout the patient's life. 
The complete, regular follow-ups in our study enabled the collection of information to analyze the progression of involution. In our report, acute ROP not requiring treatment began to involute at a mean of 40.4 weeks postmenstrual age and finished at a mean of 50.6 weeks. The analysis of the onset of involution by zone of disease found no clinically important differences in timing, which was consistent with the results of Repka et al. 7 However, the mean onset time in our study was 1.8 weeks later than that of the the study of Repka et al. 7 This difference could be explained by differences in the follow-up pattern, as infants with posterior disease or higher stages of ROP were seen more frequently in the study of Repka et al. 7  
The duration time of ROP is a controversial issue in many articles. Eliason et al. 21 reported that the average duration of untreated ROP in white non-Hispanics was 8.6 ± 5.4 weeks, while Ju et al. 22 reported that the mean duration of ROP with spontaneous regression was 5.6 ± 3.1 weeks in China. This difference is due to the different investigators selecting different and obscure criteria to define the completion of involution, which makes it difficult to compare results. A specific definition of completion time and a regular follow-up pattern are necessary to draw more comprehensive and objective conclusions. In our study, the minimal duration time was only 1.0 week, while the maximal was 43.6 weeks, with a mean of 10.3 weeks. The duration time was influenced by many factors, such as severity of ROP, BW, and GA. Analysis by severity of ROP revealed that involution began the earliest and took the shortest time with the mildest disease, and began the latest and took the longest time with the most severe disease. Thus, the duration time of involution is directly related to the maximum stage reached. Considering the natural course of the pathognomonic lesions associated with acute ROP, 23,24 the relationship between the duration time and severity of ROP is understandable. Our analysis by BW and GA showed that infants with BWs lower than 1000 g and GAs younger than 28 weeks had longer involution duration times than the other groups did. Generally speaking, BW and GA correlated well with the maturity of infants. From this point of view, our data reconfirmed the speculation raised by the CRYO-ROP Cooperative Group and Repka et al., 7 in which they thought that the development of acute ROP was controlled by a predetermined maturational sequence. 7,13  
Although extensive studies have been conducted regarding risk factors for severe ROP or the progression of ROP, 912 little or no formal study has been conducted on the factors affecting the involution of ROP. The only study we could identify was carried out by Ju et al., 22 who indicated that retinal hemorrhage was weakly inversely associated with spontaneous regression. In our study, 14 factors were analyzed by multivariable logistic regression analysis, and three factors, CPAP, active stage 3 disease, and anemia, were identified as independent risk factors associated with delayed involution of ROP. The average duration time of involution for stage 3 disease was 23.8 weeks, which was significantly longer than the duration times of stage 2 and stage 1. It is well known that active stage 3 is characterized by the addition of extraretinal fibrovascular tissue proliferating from the former stage 2 ridge. The involution of active stage 3 disease is a comprehensive progression, with occlusion of neovascularization and subsequent retinal vascular remodeling. 25 This process explains why involution of stage 3 disease takes a long time. The CPAP and anemia are well-known risk factors for severe ROP or progression of ROP. 912,26 Our study demonstrated they also are independent risk factors for delayed involution of ROP. We postulated that the longer duration of CPAP in ROP cases might reflect the severity of apnea and episodes of tissue hypoxia, which is similar to the influence of anemia. Sustained hypoxia would impede normal retinal vascularization and prolong the period of involution. 27,28  
One major limitation of this study is the small number of infants with stage 1 and stage 3 disease. As demonstrated in our study, the duration of the involution of stage 1 disease is very short, with a mean time of 2.03 weeks. Many infants were excluded from our study because signs of regression of mild ROP were already evident at their first visit to our hospital, which made it difficult to determine the starting point of involution. The small number of stage 3 disease cases in our study might be due to the significantly reduced incidence of severe ROP after the Chinese government issued guidelines on oxygen use and promoted appropriate screening programs. Another limitation of this study is the difficulty in determining the precise time of the start and endpoints of involution. These times were derived from the average of ages at two visits, which would artificially decrease the duration measurement. 
In conclusion, this study demonstrated that the natural involution of acute ROP not requiring treatment correlated well with the severity of ROP. Most infants with stage 1 and stage 2 disease showed signs of involution before 44 weeks and finished involution before 50 weeks postmenstrual age. In such cases, it is reasonable to decrease the frequency of examinations. However, the natural involution process of stage 3 disease was significantly prolonged and always combined with cicatricial sequelae, such as vitreous condensation and vitreoretinal interface changes. Considering the special risk of developing retinal breaks and detachment, it is suggested that those infants should undergo serial examinations for a long time. Finally, the analysis of risk factors for delayed involution of ROP should be a great help to clinicians by raising awareness of the presence of the additional risk factors when monitoring premature infants with ROP. 
Acknowledgments
Supported by research grants from National Natural Science Foundation for the Young Scholars of China (No. 30901641); Shanghai Rising-Star Program (No. 10QA1401200), Young Talents Program of Shanghai Health System (No. XYQ2011058), and Shanghai Natural Science Foundation (No. 13ZR1405900) 
Disclosure: Y.-Q. Ni, None; X. Huang, None; K. Xue, None; J. Yu, None; L. Ruan, None; H.-D. Shan, None; G.-Z. Xu, None 
References
Vinekar A Dogra MR Sangtam T Retinopathy of prematurity in Asian Indian infants weighing greater than 1250 grams at birth: Ten year data from a tertiary care center in a developing country. Indian J Ophthalmol . 2007; 55: 331–336. [CrossRef] [PubMed]
Gilbert C Field A Gordillo l, et al. Characteristics of infants with severe ROP in countries with low, moderate, and high levels of development: implications for screening programs. Pediatrics . 2005; 115: e518–e522. [CrossRef] [PubMed]
Gilbert C Rahi J Eckstein M Retinopathy of prematurity in middle-income countries. Lancet . 1997; 350: 12–14. [CrossRef] [PubMed]
Xu Y Zhou X Zhang Q Screening for retinopathy of prematurity in China: a neonatal units based prospective study. Invest Ophthalmol Vis Sci . 2013; 54: 8229–8236. [CrossRef] [PubMed]
Chen Y Li X. Characteristics of severe retinopathy of prematurity patients in China: a repeat of the first epidemic? Br J Ophthalmol . 2006; 90: 268–271. [CrossRef] [PubMed]
Chinese Medical Association. Guidelines on oxygenation policies and on prevention and treatment of retinopathy of prematurity. Zhonghua Yan Ke Za Zhi . 2005; 41: 375–376.
Repka MX Palmer EA Tung B. Involution of retinopathy of prematurity. Cryotherapy for Retinopathy of Prematurity Cooperative Group. Arch Ophthalmol . 2000; 118: 645–649. [CrossRef] [PubMed]
Cryotherapy for Retinopathy of Prematurity Cooperative Group. The natural ocular outcome of premature birth and retinopathy: status at 1 year. Arch Ophthalmol . 1994; 112: 903–912. [CrossRef] [PubMed]
Seiberth V Linderkamp O. Risk factors in retinopathy of prematurity. a multivariate statistical analysis. Ophthalmologica . 2000; 214: 131–135. [CrossRef] [PubMed]
Liu PM Fang PC Huang CB Risk factors of retinopathy of prematurity in premature infants weighing less than 1600 g. Am J Perinatol . 2005; 22: 115–120. [CrossRef] [PubMed]
Bayat-Mokhtari M Pishva N Attarzadeh A Hosseini H Pourarian S. Incidence and risk factors of retinopathy of prematurity among preterm infants in Shiraz/Iran. Iran J Pediatr . 2010; 20: 303–307. [PubMed]
Chen Y Li XX Yin H Survey Group. Risk factors for retinopathy of prematurity in six neonatal intensive care units in Beijing, China. Br J Ophthalmol . 2008; 92: 326–330. [CrossRef] [PubMed]
International Committee for the Classification of Retinopathy of Prematurity, The International Classification of Retinopathy of Prematurity revisited. Arch Ophthalmol . 2005; 1237: 991–999.
Cryotherapy for Retinopathy of Prematurity Cooperative Group. Multicenter trial of cryotherapy for retinopathy of prematurity. Snellen visual acuity and structural outcome at 5 1/2 years after randomization. Arch Ophthalmol . 1996; 114: 417–424. [CrossRef] [PubMed]
Schulenburg WE Prendiville A Ohri R. Natural history of retinopathy of prematurity. Br J Ophthalmol . 1987; 71: 837–843. [CrossRef] [PubMed]
Knight-Nanan DM Algawi K Bowell R O'Keefe M. Advanced cicatricial retinopathy --outcome and complications. Br J Ophthalmol . 1996; 80: 343–345. [CrossRef] [PubMed]
Ruth A Hutchinson AK. Baker Hubbard G. Late vitreous hemorrhage in patients with regressed retinopathy of prematurity. J AAPOS . 2008; 12: 181–185. [CrossRef] [PubMed]
Droste PJ Trese MT. Measurement of low vision in advanced cicatricial retinopathy of prematurity. J Pediatr Ophthalmol Strabismus . 1995; 32: 188–190. [PubMed]
Park KH Hwang JM Choi MY Yu YS Chung H. Retinal detachment of regressed retinopathy of prematurity in children aged 2 to 15 years. Retina . 2004; 24: 368–375. [CrossRef] [PubMed]
Jandeck C Kellner U. Foerster MH.Late retinal detachment in patients born prematurely: outcome of primary pars plana vitrectomy. Arch Ophthalmol . 2004; 122: 61–64. [CrossRef] [PubMed]
Eliason KJ Dane OJ Amsel E Richards SC. Incidence, progression, and duration of retinopathy of prematurity in Hispanic and white non-Hispanic infants. J AAPOS . 2007; 11: 447–451. [CrossRef] [PubMed]
Ju RH Zhang JQ Ke XY Spontaneous regression of retinopathy of prematurity: incidence and predictive factors. Int J Ophthalmo . 2013; 6: 475–480.
Flynn JT. Acute proliferative retrolental fibroplasia: evolution of the lesion. Albrecht Von Graefes Arch Klin Exp Ophthalmol . 1975; 195: 101–111. [CrossRef] [PubMed]
Flynn JT Bancalari E Bachynski BN Retinopathy of prematurity. Diagnosis, severity, and natural history. Ophthalmology . 1987; 94: 620–629. [CrossRef] [PubMed]
Benjamin LE Hemo I Keshet E. A plasticity window for blood vessel remodelling is defined by pericyte coverage of the preformed endothelial network and is regulated by PDGF-B and VEGF. Development . 1998; 125: 1591–1598. [PubMed]
Graziano RM Leone CR Cunha SL Pinheiro AC. Prevalence of retinopathy of prematurity in very low birth weight infants. J Pediatr (Rio J) . 1997; 73: 377–382. [CrossRef] [PubMed]
Gariano RF Gardner TW Retinal angiogenesis in development and disease. Nature . 2005; 438: 960–966. [CrossRef] [PubMed]
Arjamaa O. Nikinmaa M. Oxygen-dependent diseases in the retina: role of hypoxia-inducible factors. Exp Eye Res . 2006; 83: 473–483. [CrossRef] [PubMed]
Figure
 
Fundus photograph to illustrate the natural involution of mild acute ROP. (AC) Involution of stage 1 disease. View of the demarcation line (white arrows) which gradually blurred and disappeared by 42 weeks postmenstrual age. (DI) Involution of stage 2 disease. The ridge (white arrows) at the junction between vascularized and avascular retina was the hallmark of stage 2 ROP. Small isolated tufts, commonly called “pop-corn” (black arrows), were seen posterior to the ridge. During the involution process, the elevated ridge flattened, the “pop-corn” disappeared, and the retinal vessels grew into a more peripheral zone. Vascularization of the retina was completed by 49 weeks postmenstrual age. (JO) Involution of mild stage 3 disease. A stage 2 disease (ridge) was found at 39 weeks postmenstrual age and progressed to stage 3 at 40 weeks with neovascularization extended from the ridge into the vitreous in the inferior temporal region (black arrow). Note the tortuosity and dilatation of posterior pole vessels that were insufficient for plus disease, which we called “preplus” disease (red arrows). The extraretinal fibrovascular tissue regressed spontaneously after 2 weeks with a flattening ridge. The retinal vessels stepped over the region that previously had acute ROP and grew into a more peripheral zone at 62 weeks postmenstrual age. However, vitreous condensation and vitreous membrane (yellow arrows) were seen in the area of previously active stage 3 disease.
Figure
 
Fundus photograph to illustrate the natural involution of mild acute ROP. (AC) Involution of stage 1 disease. View of the demarcation line (white arrows) which gradually blurred and disappeared by 42 weeks postmenstrual age. (DI) Involution of stage 2 disease. The ridge (white arrows) at the junction between vascularized and avascular retina was the hallmark of stage 2 ROP. Small isolated tufts, commonly called “pop-corn” (black arrows), were seen posterior to the ridge. During the involution process, the elevated ridge flattened, the “pop-corn” disappeared, and the retinal vessels grew into a more peripheral zone. Vascularization of the retina was completed by 49 weeks postmenstrual age. (JO) Involution of mild stage 3 disease. A stage 2 disease (ridge) was found at 39 weeks postmenstrual age and progressed to stage 3 at 40 weeks with neovascularization extended from the ridge into the vitreous in the inferior temporal region (black arrow). Note the tortuosity and dilatation of posterior pole vessels that were insufficient for plus disease, which we called “preplus” disease (red arrows). The extraretinal fibrovascular tissue regressed spontaneously after 2 weeks with a flattening ridge. The retinal vessels stepped over the region that previously had acute ROP and grew into a more peripheral zone at 62 weeks postmenstrual age. However, vitreous condensation and vitreous membrane (yellow arrows) were seen in the area of previously active stage 3 disease.
Table 1
 
Start Point of Involution by Zone of Acute ROP
Table 1
 
Start Point of Involution by Zone of Acute ROP
Zone No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
II 21 40.6 (37.1–45.1)
III 61 40.3 (36.3–44.4)
Total 82 40.4 (36.5–44.4)
Table 2
 
Start Point of Involution by Stage of ROP
Table 2
 
Start Point of Involution by Stage of ROP
Stage No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
1 11 38.1 (34.5–39.6)
2 54 40.3 (37.2–44.6)
3 17 42.3 (39.6–45.2)
Total 82 40.4 (36.5–44.4)
Table 3
 
End Point of Involution by Zone of Acute ROP
Table 3
 
End Point of Involution by Zone of Acute ROP
Zone No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
II 21 56.6 (39.2–84.2)
III 61 48.5 (37.5–70.4)
Total 82 50.6 (39.0–75.0)
Table 4
 
End Point of Involution by Stage of Acute ROP
Table 4
 
End Point of Involution by Stage of Acute ROP
Stage No. Eyes Mean Postmenstrual Age, wks (5%–95%)*
1 11 39.6 (36.5–42.2)
2 54 48.3 (40.6–59.5)
3 17 66.0 (48.4–84.1)
Total 79 50.6 (39.0–75.0)
Table 5
 
Univariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Table 5
 
Univariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Risk Factor β OR 95% CI of OR P Value
CPAP 3.43 30.89 8.84∼107.95 <0.0001
NRDS 3.23 25.19 7.59∼83.62 <0.0001
Preplus disease 3.08 21.86 2.60∼183.48 0.004
Sepsis 2.74 15.52 1.80∼133.56 0.01
Active stage 3 disease < 5 clock-h 2.66 14.30 3.64∼56.17 <0.0001
Oxygen 2.48 12 2.59∼55.68 0.002
Anemia 2.29 9.87 3.36∼28.97 <0.0001
Acidosis 1.69 5.44 1.67∼17.78 0.005
Zone II disease 1.44 4.21 1.48∼11.94 0.006
Male 1.30 3.67 1.21∼11.09 0.02
GA < 28 wks 1.41 4.08 0.94∼17.75.20 0.06
Asphyxia 0.85 2.34 0.75∼7.28 0.14
Hyperbilirubinaemia 0.42 1.53 0.62∼3.77 0.36
BW < 1000 g 0.38 1.11 0.99∼1.25 0.08
Table 6
 
Multivariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Table 6
 
Multivariable Logistic Regression Models of Potential Risk Factors Associated With Delayed Regression of Mild Acute ROP
Risk Factor β OR 95% CI of OR P Value
CPAP 3.07 20.50 4.55∼101.60 <0.0001
NRDS 1.04 2.84 0.22∼37.14 0.43
Preplus disease 1.15 3.17 0.10∼99.75 0.51
Sepsis −0.78 0.46 0.03∼7.89 0.59
Active stage 3 disease < 5 clock-h 2.81 16.60 2.27∼121.09 0.006
Oxygen 0.89 2.44 0.33∼18.07 0.38
Anemia 1.61 4.99 1.12∼22.29 0.03
Acidosis 0.97 2.64 0.44∼15.83 0.29
Zone II disease 1.45 4.28 0.76∼24.02 0.09
Male 0.84 2.32 0.39∼13.86 0.35
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×