Adults born preterm: a review of general health and system-specific outcomes

General comments on outcomes

Tables 1-6 provide summaries from 126 publications, 12 of which were systematic reviews and 114 were cohort studies on organ/system-specific outcomes. Seventy-three of 114 cohort studies (64%) included concurrent controls. Many large national or regional cohorts cited in the tables were from Australia, France, Finland, Germany (Bavaria), the Netherlands, New Zealand, Norway and Sweden, or from high-risk infant follow-up cohorts from Canada, the United Kingdom and the United States.

While it is not easy to provide a uniform assessment of the strengths and limitations of each of the selected study, we wish to underscore that studies using very large national databases [14-21] clearly had an advantage in offering more stable estimates of adult outcomes.

The cohort studies are listed in the tables based on the age of the oldest participant in the report in an ascending order. We focused on adolescent and adult outcomes but retained eight reports [22-29] that included participants 10 years of age or younger because the findings seemed complementary to explaining later outcomes. In 75 of 114 (67%) of the cohort studies, the oldest participants were 20–39 years of age. Because of this, it is noteworthy that a large majority of participants were those born between the mid-to-late 1970s through the 1980s. In only six reports were individuals born before 1969 included [14, 21, 30-33]. Participants in 78 of 114 (68%) cohort studies were extremely or moderately premature at birth (22–28 weeks or 32–34 weeks, respectively) and/or were extremely low or very low birthweight. The implications of these observations are discussed later.

Preterm, Small for Gestational Age (SGA) and Intrauterine Growth Restriction (IUGR)

In 9 of 126 reports, SGA cohorts were included [30, 34-41], and in two others, IUGR cohorts were included [22, 23]. These reports also contained outcomes for adults born preterm who were not SGA. And, one study [30] reported outcomes for adults born term and SGA, compared to those born just preterm (n = 2931), and those who were low birthweight (n = 2176). We acknowledge that the so called ‘pure’ preterm phenotype may be different from the IUGR/SGA phenotypes, even when both coexisted. However, we retained these 11 reports because removing them did not seem to alter our conclusions, but more importantly, might have introduced a bias.

Sibling rank and multiple preterm infants in single households

Additional factors that could affect long-term outcome could be sibling rank-order, and whether a preterm infant is one among other preterm infants in the same household. These variables can provide additional explanations about the role of genetic, environmental and lifestyle factors on outcomes. However, there were no reports exploring these variables.

In the section that follows, we provide a summary of individual organ/system-specific outcomes.

Brain structure, neurosensory and cognitive outcomes

In Table 1, we have summarised 28 reports dealing with the brain structure, neurosensory and cognitive outcomes. Four of the 28 were systematic reviews and 24 cohort studies. The most consistent observation from brain imaging studies was varying combinations of reduced volumes for the whole brain, or in specific regions of the brain, namely a loss of grey and white matter in the temporal, frontal, insular and occipital areas, thalamus, caudate nucleus, putamen, globus pallidus and corpus callosum [42-45]. A systematic review focused on outcomes following neonatal cerebellar injury and found reduced cerebellar volumes were correlated with higher prevalence of behavioural abnormalities [46]. Other findings included evidence of persistent corticospinal tract injury [47], and widespread reductions and disruptions in intrinsic and resting-stage connectivity of networks [23, 42, 48-54]. Reduced overall IQ scores [55-59] and lower math abilities [14, 60] were also reported frequently.

Some studies showed correlations between abnormalities in brain structure and volume with reduced IQ [42, 44] and impaired executive functions [48, 49, 61], visual short-term memory, working memory and other socio-cognitive functions [49, 50]. However, the correlations between abnormal brain structure and functional outcomes were inconsistent, perhaps because of the differences in cohort characteristics such as gestational age, birthweight, age at follow-up, and attrition rates, or due to difference in outcome assessment methods such as brain imaging techniques and neurosensory and functional evaluations.

Also, there were studies that reported the remarkable ability of the human brain to reorganise its anatomical connections based on the functional needs [22, 52]. Such plasticity probably explains how a large proportion of adults born preterm, even after surviving neonatal brain injury, continue to function as normally as their term-born counterparts.

The factors worsening poor outcomes include the degree of prematurity [13, 36], exposure to postnatal dexamethasone or other postnatal steroids [25] and intrauterine growth restriction [22]. No study prospectively investigated the potentials for protective effects from early developmental interventions, the role for preventive and therapeutic neuro-rehabilitation, the positive effects from optimal nurturing and care at home or the role of optimal neonatal and postnatal nutrition on improving the structure and functions of the neurosensory and cognitive systems.

Mental health and psychiatric outcomes

In Table 2, we have summarised 19 reports related to mental health and psychiatric outcomes, two of which were systematic reviews and 17 cohort studies. A systematic review of five studies reporting on ‘any psychiatric diagnosis’ among 565 preterm and 533 control individuals reported that the risk of anxiety/depression were nearly fourfold higher among adults born preterm or low birthweight [62]. Another review [63] and a few cohort studies reported higher frequencies of school problems, anxious and depressive symptoms while transitioning from childhood to adolescence [64], risk of hospitalisation for mental disorders [38] and borderline behavioural problems [58].

A consistent observation was that adults born preterm had different, not necessarily abnormal, personality types compared to their term-born counterparts. These included possessing less negative emotions, being more ‘dutiful’, ‘cautious’ and ‘shy’ [65]. They were also less prone to criminal and risk-taking behaviours, smoking and illicit drug and alcohol abuse [66, 67]. Being timid, withdrawn and shy might explain their social difficulties in adult roles, such as peer/partner romantic relationships [68]. In addition, they had higher prevalence of neuroticism and withdrawal features, and autistic behaviours [68, 69]. Such personality features are probably responsible for this group being at a higher risk for peer-victimisation and bullying [24, 70, 71]. Boys and men tended to fare worse compared to girls and women [38, 58], and IUGR and SGA were added risk factors for poor psychosocial status [37, 38, 72].

Cardiovascular outcomes

Cardiovascular outcomes in adults born preterm are summarised in Table 3 from 3 systematic reviews and 21 cohort studies. One consistent observation was that adults born preterm had significantly higher systolic and diastolic blood pressure values, ranging between 2 and 8 mmHg, compared to their age-matched term controls [40, 73-75]. A significantly higher prevalence of hypertension was also reported in one study [39], but this finding was not consistent. A report from the Adults Born Preterm International Coalition provided precise estimates of blood pressure changes. This was a large patient-level meta-analysis using raw data from nine longitudinal cohorts from Australia, Canada (two sites), Finland, Ireland, the Netherlands, Norway, and the USA (two sites) [74]. A pooled sample of 1571 adults who were very low-birthweight infants was compared with 777 control adults with normal birthweight. After adjustment for age, sex, and the study site, adults born very low birthweight had 3.4 mmHg (95% confidence interval, 2.2–4.6) higher systolic and 2.1 mmHg (95% CI, 1.3–3.0) higher diastolic pressure. The difference in mean systolic pressure in men was 1.8 mmHg (95% CI, 0.1–3.5) and in women 4.7 mmHg (95% CI, 3.2–6.3). The biological basis for higher risk for increased blood pressure in adult women compared to adult men remains to be studied.

Abnormal growth of the vascular tree and persisting endothelial malfunction following preterm birth and/or low birthweight could be the pathogenesis for higher blood pressure in adults born preterm [76]. A series of studies from a single group of investigators has reported abnormalities in different components of the vascular architecture, and in the shape and contractive functions of heart chambers in children, adolescents and young adults born preterm [77-82]. A German cohort study reported significant abnormalities in carotid artery thickness and microvascular density in children (mean age 11.8 years) born extremely preterm [28]. However, a Swedish study based on 56 school children, 39 of whom were very preterm at birth (mean gestational age, 29 weeks) and 17 term controls, reported no differences in the carotid artery elasticity and structure between the two groups [35]. But, taken together, these findings inform us that following preterm birth there is a tendency for abnormal vascular architecture causing increased intimal and medial thickness and poor distensibility of large blood vessels, increased right and left ventricular muscle mass with reduced distensibility, as well as peripheral vascular abnormalities. These observations could imply a higher risk for earlier onset of cardiovascular disorders in adults born preterm. Maternal pre-eclampsia and hypertension further enhanced the risk for higher blood pressure in adults [83, 84].

The information concerning the risk for coronary heart disease and stroke among adults born preterm was not consistent across studies. A report based on the Helsinki Birth Cohort (participants born between 1924 and 1944) [21] found no increased risk of coronary heart disease or stroke among old age adults born preterm, although women born early preterm had higher rates of coronary heart disease. There was a discrepancy between increased risk factors in younger generations born preterm and little or no increase in manifest disease in older age. However, a Swedish cohort study (participants born between 1925 and 1949) [30] found that compared to adults who had normal birthweight, those born small for gestational age were at increased risk for ischaemic heart disease (adjusted hazard ratio, 1.64; 95% CI, 1.23 to 2.18). This effect, which was independent of gestational age, suggests poor foetal growth may be a risk factor.

Despite the inconsistency in the rates of manifest cardiovascular diseases and stroke, adults born preterm had a much higher prevalence of cardio-metabolic risk factors, which could portend higher or earlier onset of cardiovascular diseases and stroke. These abnormalities included dysglycaemia, abnormal lipid profiles and abnormal distribution of body fat [19, 85-87].

Pulmonary function and lung diseases

In Table 4, we have summarised the pulmonary outcomes of adults born preterm from 18 reports, two of which were systematic reviews and 16 cohort studies. Another review on the lung consequences of adults born preterm provided descriptive outcomes [88]. As this report did not provide search history results, we have not included this review in the table. A systematic review of 21 publications summarised that preterm-born adults had significantly decreased per cent forced expiratory volume by 1 second, which improved following a single dose bronchodilator, and the response was larger in those who had suffered from bronchopulmonary dysplasia (BPD) during the neonatal period [89]. Other reports corroborate the increased airway resistance [90, 91], decreased exercise tolerance [92-94] and reduced carbon monoxide diffusion capacity [95] among adults born preterm. However, abnormal findings were not uniform. Lung function abnormalities were either mild [96], or did not exist during maximum exercise [27], and in some cases, despite lower muscular fitness, the participants felt themselves to be fit [97].

General and respiratory health and the frequency of respiratory symptoms were worse among adults who had BPD in the neonatal period [98, 99]. In one study, the prevalence of childhood asthma was inversely related to gestational age at birth [20]. Moderately preterm infants were 3.9 times more likely to report a diagnosis of asthma in childhood compared to their term-born counterparts.

Another noteworthy finding was a changing outcome profile for infants born in different epochs. A comparison of pulmonary function from a cohort born in 1991–1992 with those born in 1999–2000 showed that the outcome was much better for the latter. The investigators concluded that this was probably due to more widespread use of antenatal corticosteroid, and improving quality of ventilatory care during the neonatal period [100].

Despite many publications on pulmonary outcomes, we do not know whether preterm birth leads to an accelerated age-related decline in lung functions, and how adverse influences such as smoking, indoor and outdoor air pollution and exposures to occupational air pollutions affect the pulmonary function trajectory during ageing.

Metabolic and kidney outcomes

In Table 5, we summarise the findings from 13 reports, on metabolic and kidney outcomes, one of which was a systematic review and 12 cohort studies. A consistent observation was a higher prevalence of the precursors for type 2 diabetes in early adulthood. These included higher insulin resistance or lower insulin reserve, higher frequencies of dysglycaemia, abnormal lipid profiles, and abnormal distribution of body fat with a tendency for higher abdominal fat [101-104]. In a population-based cohort study, the rates of prescriptions filled for diabetic medications were slightly higher for those born preterm compared to their term controls [105].

Preterm birth and intensive care can have significantly higher adverse effects on the developing kidney. Approximately 60% of nearly 1 million adult nephrons are formed during the second and third trimester of foetal life [106, 107]. Thus, preterm birth can prevent or retard nephrogenesis, and the nephrotoxic medications administered in the intensive care can have significant negative effect on postnatal nephrogenesis, potentially increasing the risk for chronic kidney disease later in life. Yet, very few studies have examined long-term renal status of adults born preterm. Over an 18-years study period, 20 extremely low-birthweight infants with a history of acute renal failure in the neonatal period were followed [108]. At a mean age of 7.5 years (range 3.2–18.5 years), the renal function deteriorated in nine patients with increasing proteinuria. Other indicators of progression of renal malfunction were a high urinary to plasma creatinine (>0.6) at one year of age, and a tendency to develop obesity with body mass index >85th percentile.

An overview of kidney-associated outcomes in relation to foetal growth, preterm birth and postnatal factors summarised besides prematurity and small for gestational age high birthweight and gestational exposure to maternal diabetes or obesity were also associated with an increased risk of hypertension and kidney disease in later life [109]. The authors further concluded that adverse maternal nutrition before and during pregnancy affected foetal kidney development, as did infant's postnatal nutrition. Nephrogenesis will also be adversely affected following neonatal acute kidney injury, common after preterm birth and clinical events causing systemic hypoxia and kidney ischaemia [109].

More longitudinal studies need to focus on kidney outcomes. They are needed to understand the negative effects of nephrotoxic medications on long-term kidney health. Similarly, we need to understand the role of adequate macro- and micronutrients in restoring renal health among adults born at extremely low gestational ages – an increasingly large segment of our population.

Miscellaneous outcomes

In Table 6, we have summarised the findings 24 reports dealing with miscellaneous outcomes. Two of these were systematic reviews on quality of life and 22 were cohort studies covering outcomes, such as the quality of life, growth, social skills, sleep, physical activity, eating behaviour, gonadal function, adult financial status (wealth) and overall mortality.

A study based on a large regional cohort reported that after adjusting for socio-economic factors and other confounders, VLBW adults had significantly reduced earning capacity and wealth [14]. These findings were interpreted to be the consequence of reduced math and language skills, lower IQ and difficult social skills among very low-birthweight adults.

Two studies on the biology of ageing reported reduced telomere length among preterm-born adolescents and adults [110, 111]. In one of these studies [110], a group of college students, masked to the clinical participants' medical history, assessed the ‘facial age’ of preterm-born adults and controls. Adults born preterm tended to ‘look’ much older than their chronological age and these results correlated with the extent of reduced telomere length. These findings indicate that preterm birth may accelerate the ‘ageing clock’. However, there are many unknown issues related to the biology of ageing and preterm birth before we can understand the cause and effect of gestational age on the ageing process [112].