The terms ventriculomegaly and hydrocephalus are often used inconsistently and imprecisely in the literature. In this discussion we use ventriculomegaly as an overarching term to describe excessive cerebrospinal fluid (CSF) within enlarged cerebral ventricles without reference to etiopathogenesis, natural history, or outcome. Ventriculomegaly may be subcategorized in a number of ways, based on the underlying pathogenetic mechanism, clinical course, and associated features ( Fig. 3.1 ). Hydrocephalus is one major category of ventriculomegaly that results from distention of the ventricles by CSF that accumulates because of an imbalance in CSF production and absorption. Ex vacuo ventriculomegaly is a second major category that is not related to the hydrocephalic state (see later and Fig. 3.1 ). This discussion addresses those forms of ventriculomegaly-hydrocephalus that have their mechanistic origin in the fetal period , even when development of hydrocephalus is delayed until after birth, for example in some cases of myelomeningocele and the Dandy-Walker malformation. Conversely, causes of neonatal ventriculomegaly-hydrocephalus with postnatal origins, such as neonatal meningitis and intracranial hemorrhage, are discussed in subsequent chapters.
Fetal Ventriculomegaly
Understanding of the prevalence, natural history, and outcome of fetal ventriculomegaly has advanced significantly in recent decades. This advance is in large part due to widespread implementation in developed countries of standardized ultrasound screening protocols during pregnancy. As a consequence, fetal ventriculomegaly is now identified in up to 1% of all pregnancies, making it the leading referral diagnosis for fetal neurological consultation in larger clinical programs.
Classification of Fetal Ventriculomegaly
Fetal ventriculomegaly has been classified by a number of different criteria ( Box 3.1 ). These criteria include severity (mild, moderate, or severe); evolution (transient, stable, or progressive); laterality (unilateral or bilateral); symmetry (symmetrical or asymmetrical); and association with other findings (isolated or complex).
Severity of Fetal Ventriculomegaly
This is broadly classified as mild (≤15 mm) or severe (> 15 mm), with some authors further categorizing the milder form into mild (10 to 12 mm) and moderate (13 to 15 mm) categories. Milder forms of ventriculomegaly (bilateral, unilateral, or asymmetrical) are far more common than severe forms and have a reported prevalence that varies widely from 1 to 22/1000 births. Conversely, severe ventriculomegaly is uncommon. Although, in one large study with an overall fetal ventriculomegaly incidence of 3.8/1000 births, the incidence of severe ventriculomegaly was 2/1000 births. In some studies the percentage of fetal ventriculomegaly in the severe range (>15 mm) was as high as 53% to 55%. Mild isolated ventriculomegaly is unilateral in 28% to 33% of cases. Severe unilateral ventriculomegaly is rare and suggests either occlusion of the foramen of Monro or hemispheric tissue loss. One large recent study of 432 cases included a “gray zone” group with ventricles between 7 and 10 mm, and found that only 2.8% of these cases progressed to ventriculomegaly (>10 mm).
Evolution of Fetal Ventriculomegaly ( Box 3.2 )
The terms stable, transient, or progressive have been used to describe the evolution of ventriculomegaly across gestation. The evolution of ventriculomegaly (i.e., the rates of progression, stabilization, and resolution) during fetal life varies widely in different reports . The most common scenario is that of milder forms of ventriculomegaly, often unilateral or asymmetrical, that remain stable or are transient, normalizing on follow-up studies. Among cases of isolated mild-moderate ventriculomegaly (<15 mm) a small minority (11% to 16%) progress to the severe category, while 43% to 75% stabilize, and 14% to 62% normalize their ventricular size. In the mild-moderate group of isolated ventriculomegaly, reported spontaneous resolution rates range from 53% to 64% in the 10- to 12-mm group and from 6% to 47% in the 13- to 15-mm category. Conversely, severe ventriculomegaly (>15 mm) is rare and has a low likelihood of spontaneous resolution. The unusual cases of progressive ventriculomegaly are usually severe at time of the standard mid-gestation anatomical survey.
Isolated Versus Complex Fetal Ventriculomegaly ( Box 3.3 )
Isolated ventriculomegaly refers to cases with no other anomalies or risk factors and occurs in 0.39 to 0.87 per 1000 births. In most, but not all, reports the rate of associated anomalies is low when the maximum ventricular diameter is less than 15 mm. Conversely, one large review of 355 fetuses with ventriculomegaly less than 15 mm found a 55% incidence of complex ventriculomegaly, with fetal anomalies in 43% overall. Within the less than 15 mm ventriculomegaly group, the associated central nervous system (CNS) anomaly rate has ranged from 6% to 41% in the 10- to 12-mm group, and between 32% and 76% in the 13- to 15-mm group.
Ventricle Size ≤15 mm
- ◆
10–12 mm 6%–41% associated with CNS anomalies
- ◆
13–15 mm 32%–72% associated with CNS anomalies
- ◆
1%–5% associated with congenital infections
Ventricle Size >15 mm
- ◆
32%–60% associated with CNS anomalies
- ◆
10%–20% associated with congenital infections
CNS, Central nervous system.
Complex ventriculomegaly refers to cases with associated anomalies including other malformations of the fetal brain or body, chromosomal anomalies, or signs of acquired factors that may affect outcome, such as CNS infection, intracranial hemorrhage, and hypoxia-ischemia. The incidence of associated brain malformations increases with increasing ventricular size, reaching 32% to 60% when ventricles are greater than 15 mm . Most, but not all, studies report a low rate of associated anomalies when ventricles are less than 15 mm. This wide variation in the presence of associated anomalies results in large part from the differences in timing and modality of fetal imaging. Dysmorphic ventriculomegaly is a form of complex ventriculomegaly best exemplified by agenesis of the corpus callosum with its characteristic ventricular morphology, that is, retention of the tear-drop-shaped fetal configuration posteriorly (colpocephaly) and a steer-horn configuration anteriorly (see Chapter 2 ). In one large prospective study, agenesis of the corpus callosum was diagnosed in 13% of fetal ventriculomegaly cases. Of note, before 24 weeks’ gestation, the lateral ventricles may normally have a mild colpocephalic shape, although the ventricular axes should not be parallel.
Congenital infections, another type of complex ventriculomegaly, are an important prognostic consideration in fetal ventriculomegaly, especially those with brain involvement. However, the positive detection rate based on maternal serology testing is highly variable, and depends on the timing, geographical and social setting, and specific testing performed in each of the studies. The reported incidence of congenital infections in cases of severe fetal ventriculomegaly ranges from 10% to 20%, but only from 1% to 5% in mild ventriculomegaly cases. In one review, evidence of congenital infection was present in 0.4% of mild ventriculomegaly (10 to 12 mm) and 1.5% of moderate ventriculomegaly (12 to 15 mm). Cerebral ventriculomegaly identified by fetal ultrasonography (US) is present almost 20% of cases of fetal cytomegalovirus encephalitis, and approximately half the cases appear isolated ( Box 3.4 ).
Diagnosis of Fetal Ventriculomegaly: Imaging Modalities
Fetal ultrasound is the standard prenatal screening modality for identifying fetal ventriculomegaly, and has become an important tool for monitoring ventricular size, configuration, and evolution over gestation. Widespread clinical use over more than four decades has provided a wealth of experience and a large body of population-based normative data. By convention, fetal ventricular size is measured by ultrasound as the diameter of the lateral ventricular atrium on an axial view ( Fig. 3.2 ), at the level of the thalami or the glomus of the choroid plexus. Using this approach, the ventricular size remains relatively stable at around 6 to 7 mm during normal development between 14 weeks and term. A widely used criterion for diagnosing ventriculomegaly is ventricular diameter of ≥10 mm (>4 standard deviations above the mean) at the axial level discussed earlier. Fetal US can also identify other features of ventriculomegaly, such as the “dangling choroid” sign ( Fig. 3.3 ), and may help to distinguish hydrocephalic from nonhydrocephalic ventriculomegaly. Specifically, large ventricles and a large head circumference suggest fetal hydrocephalus, while large ventricles and a small head are more suggestive of ex vacuo ventriculomegaly. Fetal US may also show increased echogenicity of the ventricular lining when ependymitis results from intraventricular hemorrhage or infection. Informed prognostication in cases of ventriculomegaly (see later) is based on more than ventricular size, and includes the association with other brain lesions. The limited spatial resolution of fetal US may result in false negative studies for associated brain lesions in 10% to 40% of cases. Although fetal US may only detect the larger associated brain lesions, fetal MRI (see later and Chapter 10 ) has superior soft tissue resolution and sensitivity to smaller lesions.
Fetal MRI has played a major role in the advancement of fetal neurology as a discipline and is discussed in more detail in Chapter 10 . The primary added value of MRI over the fetal US diagnosis of ventriculomegaly is its enhanced resolution; the improved contrast allows for more precision in measurement and increases the detection of associated lesions. The superior soft tissue resolution of fetal brain MRI supports prognostication (see later) in fetal ventriculomegaly by providing important etiological data and detecting associated anomalies. In studies comparing the detection rate of associated lesions between these two imaging techniques, fetal MRI detects associated brain lesions in anywhere from 5% to 60% of cases previously diagnosed as isolated ventriculomegaly by fetal US ( Table 3.1 ). In cases where fetal MRI detects isolated ventriculomegaly, repeat MRI in the postnatal period provides an additional 10% to 28% yield for associated anomalies a
a References .
; this increase relates not only to the increased sensitivity of ex utero MRI, but also to disturbances in late gestation brain development occurring after the fetal MRI. Of note, conventional clinical fetal MRI relies heavily on subjective pattern recognition and 2-dimensional measurements, features that have left it prone to differences in interpretation. The application of more advanced quantitative MRI techniques to the fetus, will likely reduce these discrepancies. Several authors have reported on the use of three-dimensional volumetric MRI to measure ventricular volumes rather than the use of conventional two-dimensional measures. Andescavage et al. measured fetal ventricular volumes in healthy pregnancies between 18 and 39 weeks of gestation, and established normative data for all four ventricles and the extraaxial CSF spaces. These authors described a modest 1.4-fold increase in the volume of the lateral ventricles over this gestational period; interestingly, the growth trajectories differed, with the left ventricles larger during early gestation, but ultimately reaching a similar volume to the right ventricle.| INITIAL DIAGNOSIS TIMING AND MODALITY | REPEAT IMAGING TIMING AND MODALITY | INCREASED DETECTION OF ASSOCIATED ANOMALIES (REFS) |
|---|---|---|
| Second trimester US | Third trimester US | 13% (16) |
| Second trimester US | Second trimester MRI | 5%–60% (11, 38, 39) |
| Second trimester US | Third trimester MRI | Minimal increase (14) |
| Fetal MRI | Postnatal MRI | 10%–28% (11, 19, 20, 22, 29, 40) |
In summary, fetal US is a valuable tool for diagnosing and monitoring the evolution of ventriculomegaly. The superior tissue resolution provided by fetal MRI makes it the optimal technique to exclude other brain anomalies associated with fetal ventriculomegaly, which is of great importance for prognostication. Major events in cerebral cortical development occur in the late second and third trimesters (see Chapters 6 and 7 ) and are readily evaluated by fetal MRI (but not fetal US). In countries where pregnancy termination is illegal beyond mid-gestation, identifying these later cortical events by fetal MRI may not directly affect management of the pregnancy. However, this information may satisfy the parent’s need-to-know imperative and allow them to better prepare for events after birth. In one study, the rate of identifying new lesions increased by 13% when second trimester US was repeated in the third trimester. In those cases with confirmed isolated ventriculomegaly by second trimester MRI, repeat MRI studies in the third trimester provided minimal increased yield for associated anomalies; this finding also confirmed that delaying the first MRI study to the third trimester provides no significant benefit.
Differential Diagnosis of Fetal Ventriculomegaly
Ventriculomegaly is usually readily recognized by experienced ultrasonographers, but may on occasion be mistaken for other fluid-filled supratentorial lesions, such as hydranencephaly ( Fig. 3.4 ), holoprosencephaly ( Chapter 2 ), or enlarged cavum septi pellucidi or cavum velum interpositum. These issues are readily resolved by fetal MRI.
Congenital Hydrocephalus
A minority of fetal ventriculomegaly cases enter a phase of progressive ventricular enlargement (which may have arrested by the time of diagnosis) because CSF production exceeds absorption and therefore meet criteria for congenital hydrocephalus, that is, hydrocephalus overt at birth. The vast majority of cases of fetal hydrocephalus is due to decreased CSF absorption which is essentially obstructive in nature. Excessive CSF production due to a choroid plexus papilloma, for example, is a rare cause of hydrocephalus (discussed later). With implementation of fetal US screening programs, congenital hydrocephalus is now increasingly identified before birth; however, some cases are only detected in the early newborn period and are assumed to have fetal origins.
Anatomical Abnormalities in Congenital Hydrocephalus
Hydrocephalus of fetal onset may result from abnormalities anywhere in the CSF pathway from formation in the choroid plexus to absorption in the arachnoidal villi. The specific pattern of enlarged CSF spaces may suggest the site of obstructed CSF flow, allowing further categorization of hydrocephalus. Enlargement of any or all of the ventricles due to obstruction of CSF flow upstream or at the fourth ventricular foramina is termed noncommunicating hydrocephalus ; in these cases the extraaxial CSF spaces appear normal or decreased in size. Impaired CSF flow distal to the fourth ventricular foramina (e.g., in the subarachnoid spaces or arachnoid granulations) results in communicating hydrocephalus; in which case the extraaxial CSF spaces may be enlarged. In noncommunicating hydrocephalus, the location of obstructed CSF flow is most commonly at the aqueduct of Sylvius and the fourth ventricular foramina. Aqueductal stenosis ( Fig. 3.5 ) is inferred when the lateral and third ventricles are enlarged and the fourth ventricle is normal in size. Conversely, obstruction of fourth ventricular CSF egress through the foramina of Magendi and/or Luschka leads to tetraventricular hydrocephalus.
Etiology of Fetal Ventriculomegaly and Congenital Hydrocephalus
In the majority of mild isolated fetal ventriculomegaly cases, the underlying mechanism remains unknown. Some consider this a normal variant, especially when the ventriculomegaly is self-resolving. Delayed fenestration of the Blake’s pouch (see Chapter 4 ) has been suggested to underlie some of these transient forms, especially in cases with associated enlargement of the fourth ventricle. Although the etiology of congenital hydrocephalus is heterogeneous, most cases appear to result from developmental disorders of the brain and its CSF circulatory system. The causes of fetal hydrocephalus are generally similar to those of neonatal hydrocephalus, although, not surprisingly, the severity of the hydrocephalus in fetal cases tends to be greater. In addition, factors such as geographical location, the timing and modalities of brain imaging, and other testing underlie the considerable variance in the spectrum and prevalence of etiologies reported. The major causes of congenital hydrocephalus , as identified in two large series conducted in neonatal neurosurgical services, are presented in Table 3.2 . In myelomeningocele with the Chiari type II malformation (see Chapter 1 ; Fig. 3.6 ) the small and crowded posterior fossa may obstruct CSF flow at aqueductal, foraminal, and subarachnoid levels. In these studies, aqueductal stenosis (see Fig. 3.5 ) accounted for approximately one third of cases, making it the second most common cause of fetal hydrocephalus after the Chiari II malformation. The Dandy-Walker malformation (see Chapter 4 ) and communicating hydrocephalus account for most of the remaining cases of neonatal hydrocephalus. In aqueductal stenosis, the lateral and third ventricular hydrocephalus tends to be severe and progressive. In severe cases, the massive ventricular distention may have destructive effects on the septal leaflets and corpus callosum. In some cases, the overlying cerebral mantle may be stretched to the point of dehiscence, most commonly in the region of the choroidal fissure. Unlike schizencephaly, the edges of the ruptured mantle are lined with white matter. Although most examples are nonfamilial, an X-linked form of aqueductal stenosis associated with adducted thumbs and, commonly, agenesis of the corpus callosum is important to recognize because of its consistent relationship with subsequent mental retardation ( Box 3.5 ). This disorder is related to a mutation in the L1 family of neural cell adhesion molecules (L1CAM). These molecules play an important role in axon outgrowth, fasciculation, and neuronal migration. The so-called CRASH syndrome results from an L1CAM mutation and is an acronym for C orpus callosum hypoplasia, R etardation, A dducted thumbs, S pastic paraplegia, and H ydrocephalus. Additional genetic varieties of aqueductal stenosis included autosomal recessive inheritance, with a normal phenotype, and X-linked or autosomal recessive inheritance with the VACTERL association (vertebral anomalies, anal atresia, cardiovascular anomalies, tracheoesophageal fistula, renal dysplasia, limb defects). Other etiologies may lead to aqueductal stenosis, including brain stem dysgenesis (including rhombencephalosynapsis; Chapter 4 ; Fig. 3.5 ), aqueductal forking, webs, membranes, and nodules (e.g., hamartomas; Fig. 3.7 ). Approximately half of rhombencephalosynapsis cases have aqueductal stenosis (see Fig. 3.5 ), which is not surprising given the common mesencephalic origin of the vermis and midbrain (see Chapter 4 ).
| CAUSE | % |
|---|---|
| Aqueductal stenosis | 33 |
| Myelomeningocele: Chiari type II malformation | 28 |
| “Communicating” hydrocephalus | 22 |
| Dandy-Walker malformation | 7 |
| Other | 10 |
Dysgenetic
- ◆
X-linked aqueductal stenosis
- ◆
L1CAM gene mutation
- ◆
Aqueductal stenosis
- ◆
Hydrocephalus
- ◆
Adducted thumbs
- ◆
Agenesis of the corpus callosum
- ◆
Mental retardation
- ◆
Spastic paraplegia
- ◆
- ◆
Autosomal recessive aqueductal stenosis with normal phenotype
- ◆
Autosomal recessive or X-linked aqueductal stenosis
- ◆
VACTERL association (vertebral anomalies, anal atresia, cardiac anomalies, tracheoesophageal fistula, renal dysplasia, limb defects)
- ◆
- ◆
Brain stem dysgenesis (including rhombencephalosynapsis)
- ◆
Aqueductal forking, membranes, nodules (hamartomas; Fig. 3.7 ).
Figure 3.7
Midline sagittal T2-weighted magnetic resonance imaging in a case of severe congenital hydrocephalus due to aqueductal stenosis. The arrow indicates a nodule within the aqueduct blocking cerebrospinal fluid passage.
Acquired
- ◆
Fetal intraventricular hemorrhage
- ◆
Congenital infections (toxoplasmosis, CMV)
Chromosomal anomalies ( Box 3.6 ), especially aneuploidy, may be associated with fetal ventriculomegaly and are important prognostic factors. In one large study of mild ventriculomegaly (<15 mm), abnormal karyotypes (most commonly trisomy 21) were detected in 11% of cases tested antenatally, and in 14% of those tested postnatally. In the absence of structural anomalies, mild ventriculomegaly (≤15 mm) is associated with chromosomal anomalies in 3% to 10% cases. Conversely, ventriculomegaly of any size associated with other fetal anomalies has an aneuploidy rate greater than 15%. Infectious causes of fetal ventriculomegaly are prognostically important but relatively uncommon. Symptomatic newborns with known congenital CMV have a high incidence of mild-moderate ventriculomegaly ; however, ventriculomegaly is rarely the only feature of congenital CMV and is usually associated with other features such as a periventricular halo , calcifications, pseudocysts, intraventricular synechiae, and cortical and cerebellar malformations. Other infections may be associated with ventriculomegaly, including congenital Zika virus encephalopathy, and are discussed in Chapter 34 . Toxoplasmosis is a rare cause and usually results from aqueductal stenosis. The usual complication of fetal parvovirus B19 infection is fetal anemia, but mild ventriculomegaly has been described. Choroid plexus papillomas are extremely rare in the fetus and usually present with hydrocephalus due to excessive CSF production and potentially from foraminal obstruction.
Fetal intraventricular hemorrhage is increasingly detected during routine fetal imaging as a cause for fetal hydrocephalus. Arachnoid cysts may interfere with CSF flow with a picture that depends on its location; posterior fossa arachnoid cysts may obstruct posterior fossa CSF egress, whereas large suprasellar cysts may compress and obstruct third ventricular CSF flow. Malformations of the vein of Galen may result in noncommunicating hydrocephalus through aqueductal compression or communicating hydrocephalus through elevated cerebral venous pressure (see Chapter 37 ).
Timing of Congenital Hydrocephalus
In view of the heterogeneity of causes of congenital hydrocephalus, definition of a single time of onset of the disorder is not possible. However, the most important developmental processes in this context occur at approximately 6 to 10 weeks of gestation. Specifically, at this time, three events critical for the development of the CSF pathways occur: (1) development of the secretory epithelium in the choroid plexus, (2) perforation of the roof of the fourth ventricle, and (3) formation of the subarachnoid spaces. Impairments of the last two processes bear a relationship with the genesis of the hydrocephalus associated with the Dandy-Walker malformation and communicating hydrocephalus. The choroid plexuses develop between 6 and 7 weeks post-conceptual, starting in the 4th ventricle, but becoming most voluminous in the lateral ventricles. By 20 weeks the choroid plexus has an adult configuration.
The critical timing for hydrocephalus with holoprosencephaly is during this exact period. The key timing for development of aqueductal stenosis is probably later (i.e., between 15 and 17 weeks of gestation, the period of rapid elongation of the mesencephalon and evolution of the normal constriction of the aqueduct). Similar timing may be appropriate for the hydrocephalus associated with the Chiari II malformation and myelomeningocele, because aqueductal stenosis is a common cause of the disturbance in CSF dynamics in this condition (see Chapter 2 ). Clearly, of course, the inflammatory processes that cause intrauterine onset of hydrocephalus (e.g., toxoplasmosis, intracranial hemorrhage) develop still later as a consequence of the derangement in CSF flow or absorption resulting from the associated ependymitis and arachnoiditis.
Clinical Features of Congenital Hydrocephalus
Isolated nonprogressive ventriculomegaly is clinically silent at birth. Progressive ventriculomegaly in cases of congenital hydrocephalus is now frequently known before birth through the widespread use of fetal US. In cases of fetal-onset hydrocephalus , the ventricles dilate before 24 weeks of gestation, usually without change in biparietal diameter, and not until after 32 to 34 weeks of gestation does head growth consistently increase with progression of ventricular dilation. The most prominent additional clinical features are the occurrence of major extraneural anomalies in 40% to 50% of cases of fetal hydrocephalus, major CNS anomalies in 60% to 70%, and extraneural or CNS anomalies (or both) in 80%. These anomalies have a dominant effect on outcome (see the later prognosis section). As discussed earlier, MRI is the most sensitive imaging modality for assessment of hydrocephalus and associated CNS anomalies in the fetus, but postnatal MRI is significantly more sensitive to additional brain anomalies.
The neonatal presentation of congenital hydrocephalus (i.e., markedly enlarged head, full anterior fontanelle, and separated cranial sutures) is similar to, although more severe than, that described for the later postnatal development of hydrocephalus with myelomeningocele (see Chapter 1 ). Careful neonatal assessment should be made for signs of specific etiological types of congenital hydrocephalus, such as the flexion deformity of the thumbs (which can often be detected by fetal US) characteristic of approximately 50% of cases of X-linked aqueductal stenosis, the occipital cranial prominence of the Dandy-Walker formation, and the chorioretinitis of intrauterine infection by toxoplasmosis or cytomegalovirus. Serial assessments of neurological status, rate of head growth, signs of increased intracranial pressure, and ventricular size (by cranial US) are of particular value in documenting the severity and rapidity of progression of the hydrocephalic process. Earlier reports noted an incidence of one or more major CNS anomalies in 84%, and non-CNS anomalies in 56%, of congenital hydrocephalus cases. A careful assessment of the cerebral parenchyma by MRI is valuable for detection of the size of the cerebral mantle, associated anomalies of the cerebrum (e.g., disorder of neuronal migration), evidence of parenchymal destruction (e.g., calcification, cysts), and likely sites of disturbance of CSF dynamics (based on topographical distribution of ventricular dilation as described earlier).
Management of Congenital Hydrocephalus
Management of infants with congenital hydrocephalus should begin, as in other infants with major neurological diseases, with the same question: Could this problem have been prevented? Unfortunately, in the large majority of cases the answer to this question remains no.
Prenatal Management of Congenital Hydrocephalus
As indicated earlier, many cases of congenital hydrocephalus in developed countries are now identified during the fetal period. In these cases of progressive fetal ventricular dilation, rational management of fetal hydrocephalus requires an understanding of the natural history. In an earlier study of 47 infants with fetal ventriculomegaly more than half the cases died in utero, primarily because of elective termination provoked by the finding of serious neural or extraneural anomalies or both. Among the survivors (more than half of which underwent CSF shunts) all infants with subsequent neurological deficits had additional brain anomalies. These findings suggest a limited role for intrauterine management of ventriculomegaly. This principle is illustrated by the more recent observations summarized in Table 3.3 .
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
